cfahlgren1/open-inference-pricing · Datasets at Hugging Face

name stringlengths 8 58	hf_id stringlengths 12 49 ⌀	author stringlengths 3 21	providers listlengths 1 16	median_input_cost float64 0 75	median_output_cost float64 0 150	low_input_cost float64 0 75	low_output_cost float64 0 150	high_input_cost float64 0 75	high_output_cost float64 0 150	is_open_weights bool 2 classes
Qwen: QwQ 32B	Qwen/QwQ-32B	Qwen	[ { "name": "Nebius AI Studio", "context": 131000, "max_output": 131000, "input": 0.15, "output": 0.45, "latency": 0.38, "throughput": 40.22 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.18, "output": 0.2, "latency": 1.4, "throughput": 20.63 }, { "name": "Groq", "context": 131000, "max_output": 131000, "input": 0.29, "output": 0.39, "latency": 0.38, "throughput": 438.3 }, { "name": "Hyperbolic", "context": 131000, "max_output": 131000, "input": 0.4, "output": 0.4, "latency": 0.73, "throughput": 26.72 }, { "name": "Parasail", "context": 131000, "max_output": 131000, "input": 0.5, "output": 0.5, "latency": 3.22, "throughput": 32.44 }, { "name": "Nebius AI Studio (Fast)", "context": 131000, "max_output": 131000, "input": 0.5, "output": 1.5, "latency": 0.32, "throughput": 83.05 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.9, "output": 0.9, "latency": 0.82, "throughput": 124.8 } ]	0.4	0.4	0.18	0.2	0.5	1.5	true
Perplexity: R1 1776	perplexity-ai/r1-1776	perplexity-ai	[ { "name": "Perplexity", "context": 128000, "max_output": 128000, "input": 2, "output": 8, "latency": 6.35, "throughput": 85.07 } ]	2	8	2	8	2	8	true
Llama Guard 3 8B	meta-llama/Llama-Guard-3-8B	meta-llama	[ { "name": "Groq", "context": 8000, "max_output": 8000, "input": 0.2, "output": 0.2, "latency": null, "throughput": null }, { "name": "SambaNova", "context": 16000, "max_output": 16000, "input": 0.3, "output": 0.3, "latency": 0.57, "throughput": 897.3 } ]	0.3	0.3	0.2	0.2	0.3	0.3	true
Llama 3.1 Tulu 3 405B	allenai/Llama-3.1-Tulu-3-405B	allenai	[ { "name": "SambaNova", "context": 16000, "max_output": 16000, "input": 5, "output": 10, "latency": 4.53, "throughput": 48.59 } ]	5	10	5	10	5	10	true
DeepSeek: R1 Distill Llama 8B	meta-llama/Llama-3.1-8B	meta-llama	[ { "name": "NovitaAI", "context": 32000, "max_output": 32000, "input": 0.04, "output": 0.04, "latency": 1.11, "throughput": 54.37 } ]	0.04	0.04	0.04	0.04	0.04	0.04	true
AionLabs: Aion-1.0-Mini	FuseAI/FuseO1-DeepSeekR1-QwQ-SkyT1-32B-Preview	FuseAI	[ { "name": "AionLabs", "context": 33000, "max_output": 33000, "input": 0.7, "output": 1.4, "latency": 0.42, "throughput": 161.4 } ]	0.7	1.4	0.7	1.4	0.7	1.4	true
Qwen: Qwen2.5 VL 72B Instruct	Qwen/Qwen2.5-VL-72B-Instruct	Qwen	[ { "name": "Parasail", "context": 32000, "max_output": 32000, "input": 0.7, "output": 0.7, "latency": 1.78, "throughput": 15.24 } ]	0.7	0.7	0.7	0.7	0.7	0.7	true
DeepSeek: R1 Distill Qwen 1.5B	Qwen/Qwen2.5-Math-1.5B	Qwen	[ { "name": "Together", "context": 131000, "max_output": 33000, "input": 0.18, "output": 0.18, "latency": 0.26, "throughput": 394.9 } ]	0.18	0.18	0.18	0.18	0.18	0.18	true
Mistral: Mistral Small 3	mistralai/Mistral-Small-24B-Instruct-2501	mistralai	[ { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.07, "output": 0.14, "latency": 0.7, "throughput": 67.14 }, { "name": "Mistral", "context": 32000, "max_output": 32000, "input": 0.1, "output": 0.3, "latency": 0.33, "throughput": 100.8 }, { "name": "Ubicloud", "context": 33000, "max_output": 33000, "input": 0.3, "output": 0.3, "latency": 1.73, "throughput": 30.95 }, { "name": "Together", "context": 33000, "max_output": 2000, "input": 0.8, "output": 0.8, "latency": 0.47, "throughput": 80.69 }, { "name": "Fireworks", "context": 33000, "max_output": 33000, "input": 0.9, "output": 0.9, "latency": 1.14, "throughput": 33.32 } ]	0.3	0.3	0.07	0.14	0.9	0.9	true
DeepSeek: R1 Distill Qwen 32B	Qwen/Qwen2.5-32B	Qwen	[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.12, "output": 0.18, "latency": 0.32, "throughput": 43.83 }, { "name": "Cloudflare", "context": 80000, "max_output": 80000, "input": 0.5, "output": 4.88, "latency": 0.42, "throughput": 36 }, { "name": "Groq", "context": 131000, "max_output": 131000, "input": 0.79, "output": 0.79, "latency": 0.13, "throughput": 137.2 }, { "name": "NovitaAI", "context": 64000, "max_output": 64000, "input": 0.3, "output": 0.3, "latency": 8.06, "throughput": 32.05 } ]	0.79	0.79	0.12	0.18	0.5	4.88	true
DeepSeek: R1 Distill Qwen 14B	deepseek-ai/DeepSeek-R1-Distill-Qwen-14B	deepseek-ai	[ { "name": "Together", "context": 131000, "max_output": 33000, "input": 1.6, "output": 1.6, "latency": 0.32, "throughput": 163.4 }, { "name": "NovitaAI", "context": 64000, "max_output": 64000, "input": 0.15, "output": 0.15, "latency": 12.9, "throughput": 71.57 } ]	1.6	1.6	0.15	0.15	1.6	1.6	true
DeepSeek: R1 Distill Llama 70B	deepseek-ai/DeepSeek-R1-Distill-Llama-70B	deepseek-ai	[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.23, "output": 0.69, "latency": 10.86, "throughput": 84.41 }, { "name": "Nebius AI Studio", "context": 128000, "max_output": 128000, "input": 0.25, "output": 0.75, "latency": 9.23, "throughput": 64.97 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.4, "output": 0.4, "latency": 19.48, "throughput": 33.32 }, { "name": "SambaNova", "context": 16000, "max_output": 16000, "input": 0.7, "output": 1.4, "latency": 3.8, "throughput": 470.4 }, { "name": "Groq", "context": 131000, "max_output": 131000, "input": 0.75, "output": 0.99, "latency": 1.55, "throughput": 596.1 }, { "name": "Together", "context": 131000, "max_output": 16000, "input": 2, "output": 2, "latency": 5.23, "throughput": 192.7 }, { "name": "NovitaAI", "context": 32000, "max_output": 32000, "input": 0.8, "output": 0.8, "latency": 36.28, "throughput": 33.42 } ]	0.8	0.8	0.4	0.4	2	2	true
DeepSeek: R1	deepseek-ai/DeepSeek-R1	deepseek-ai	[ { "name": "Minimax", "context": 64000, "max_output": 64000, "input": 0.55, "output": 2.19, "latency": 2.41, "throughput": 20.27 }, { "name": "Nebius AI Studio", "context": 128000, "max_output": 128000, "input": 0.8, "output": 2.4, "latency": 0.43, "throughput": 12.73 }, { "name": "DeepInfra Turbo", "context": 33000, "max_output": 33000, "input": 2, "output": 6, "latency": 6.53, "throughput": 59.47 }, { "name": "inference.net", "context": 131000, "max_output": 33000, "input": 3, "output": 3, "latency": 0.76, "throughput": 38.14 }, { "name": "Parasail", "context": 128000, "max_output": 128000, "input": 3, "output": 3, "latency": 1.35, "throughput": 55.54 }, { "name": "Together", "context": 164000, "max_output": 8000, "input": 3, "output": 7, "latency": 0.87, "throughput": 58.88 }, { "name": "Friendli", "context": 164000, "max_output": 164000, "input": 3, "output": 7, "latency": 1.21, "throughput": 58.67 }, { "name": "Fireworks", "context": 164000, "max_output": 164000, "input": 3, "output": 8, "latency": 0.81, "throughput": 69.68 }, { "name": "SambaNova", "context": 8000, "max_output": 8000, "input": 5, "output": 7, "latency": 4.69, "throughput": 158.3 }, { "name": "kluster.ai", "context": 128000, "max_output": 128000, "input": 7, "output": 7, "latency": 10.41, "throughput": 34.77 }, { "name": "DeepSeek", "context": 64000, "max_output": 8000, "input": 0.55, "output": 2.19, "latency": 0.69, "throughput": 27.75 }, { "name": "DeepInfra", "context": 66000, "max_output": 8000, "input": 0.75, "output": 2.4, "latency": 2.38, "throughput": 10.05 }, { "name": "NovitaAI", "context": 64000, "max_output": 8000, "input": 4, "output": 4, "latency": 23.74, "throughput": 18.53 }, { "name": "Featherless", "context": 33000, "max_output": 4000, "input": 6.5, "output": 8, "latency": 52.12, "throughput": 19.52 }, { "name": "Avian.io", "context": 164000, "max_output": 164000, "input": 6.9, "output": 6.9, "latency": null, "throughput": null } ]	4	4	0.55	2.19	6.5	8	true
MiniMax: MiniMax-01	MiniMaxAI/MiniMax-Text-01	MiniMaxAI	[ { "name": "Minimax", "context": 1000000, "max_output": 1000000, "input": 0.2, "output": 1.1, "latency": 1.45, "throughput": 27.07 } ]	0.2	1.1	0.2	1.1	0.2	1.1	true
Microsoft: Phi 4	microsoft/phi-4	microsoft	[ { "name": "DeepInfra", "context": 16000, "max_output": 8000, "input": 0.07, "output": 0.14, "latency": 0.44, "throughput": 37.26 }, { "name": "Nebius AI Studio", "context": 16000, "max_output": 16000, "input": 0.1, "output": 0.3, "latency": 0.14, "throughput": 106 } ]	0.1	0.3	0.07	0.14	0.1	0.3	true
Sao10K: Llama 3.1 70B Hanami x1	Sao10K/L3.1-70B-Hanami-x1	Sao10K	[ { "name": "Infermatic", "context": 16000, "max_output": 16000, "input": 3, "output": 3, "latency": 1.36, "throughput": 30.45 } ]	3	3	3	3	3	3	true
DeepSeek: DeepSeek V3	deepseek-ai/DeepSeek-V3	deepseek-ai	[ { "name": "inference.net", "context": 131000, "max_output": 66000, "input": 1.2, "output": 1.2, "latency": 0.89, "throughput": 12.28 }, { "name": "Together", "context": 131000, "max_output": 12000, "input": 1.25, "output": 1.25, "latency": 0.75, "throughput": 34.84 }, { "name": "DeepSeek", "context": 64000, "max_output": 8000, "input": 0.27, "output": 1.1, "latency": 0.48, "throughput": 18.92 }, { "name": "NovitaAI", "context": 64000, "max_output": 16000, "input": 0.4, "output": 1.3, "latency": 1.43, "throughput": 34.49 }, { "name": "DeepInfra", "context": 66000, "max_output": 8000, "input": 0.49, "output": 0.89, "latency": 1.15, "throughput": 7.59 }, { "name": "Nebius AI Studio", "context": 131000, "max_output": 131000, "input": 0.5, "output": 1.5, "latency": 0.3, "throughput": 18.73 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.9, "output": 0.9, "latency": 1.11, "throughput": 49.07 } ]	0.9	0.9	0.27	1.1	1.25	1.25	true
Sao10K: Llama 3.3 Euryale 70B	Sao10K/L3.3-70B-Euryale-v2.3	Sao10K	[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.7, "output": 0.8, "latency": 0.21, "throughput": 39.36 }, { "name": "Infermatic", "context": 16000, "max_output": 16000, "input": 1.5, "output": 1.5, "latency": 0.73, "throughput": 43.27 } ]	1.5	1.5	0.7	0.8	1.5	1.5	true
EVA Llama 3.33 70B	EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1	EVA-UNIT-01	[ { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 4, "output": 6, "latency": 3.18, "throughput": 11.5 } ]	4	6	4	6	4	6	true
Meta: Llama 3.3 70B Instruct	meta-llama/Llama-3.3-70B-Instruct	meta-llama	[ { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.12, "output": 0.3, "latency": 0.88, "throughput": 14.05 }, { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.12, "output": 0.3, "latency": 0.39, "throughput": 34.73 }, { "name": "Nebius AI Studio", "context": 131000, "max_output": 131000, "input": 0.13, "output": 0.4, "latency": 0.42, "throughput": 24.18 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.4, "output": 0.4, "latency": 0.56, "throughput": 15.03 }, { "name": "Hyperbolic", "context": 131000, "max_output": 131000, "input": 0.4, "output": 0.4, "latency": 1.37, "throughput": 70.7 }, { "name": "Groq", "context": 33000, "max_output": 33000, "input": 0.59, "output": 0.79, "latency": 0.24, "throughput": 310.7 }, { "name": "Friendli", "context": 131000, "max_output": 131000, "input": 0.6, "output": 0.6, "latency": 0.41, "throughput": 103.5 }, { "name": "SambaNova", "context": 128000, "max_output": 128000, "input": 0.6, "output": 1.2, "latency": 0.9, "throughput": 131.6 }, { "name": "kluster.ai", "context": 131000, "max_output": 131000, "input": 0.7, "output": 0.7, "latency": 0.61, "throughput": 21.99 }, { "name": "Parasail", "context": 131000, "max_output": 131000, "input": 0.7, "output": 0.7, "latency": 0.64, "throughput": 66.03 }, { "name": "Cloudflare", "context": 131000, "max_output": 131000, "input": 0.75, "output": 0.75, "latency": 0.8, "throughput": 28.37 }, { "name": "Lepton", "context": 131000, "max_output": 131000, "input": 0.8, "output": 0.8, "latency": 0.39, "throughput": 47.81 }, { "name": "Together", "context": 131000, "max_output": 2000, "input": 0.88, "output": 0.88, "latency": 0.51, "throughput": 114 }, { "name": "Avian.io", "context": 131000, "max_output": 131000, "input": 0.9, "output": 0.9, "latency": 0.24, "throughput": 122.4 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.9, "output": 0.9, "latency": 0.39, "throughput": 117.1 }, { "name": "NovitaAI", "context": 131000, "max_output": 131000, "input": 0.39, "output": 0.39, "latency": 0.59, "throughput": 44.5 } ]	0.7	0.7	0.12	0.3	0.9	0.9	true
Qwen: QwQ 32B Preview	Qwen/QwQ-32B-Preview	Qwen	[ { "name": "Hyperbolic", "context": 33000, "max_output": 33000, "input": 0.2, "output": 0.2, "latency": 11.74, "throughput": 35.7 }, { "name": "Fireworks", "context": 33000, "max_output": 33000, "input": 0.9, "output": 0.9, "latency": 5.12, "throughput": 25.3 }, { "name": "Together", "context": 33000, "max_output": 33000, "input": 1.2, "output": 1.2, "latency": 15.19, "throughput": 36.67 }, { "name": "SambaNova", "context": 16000, "max_output": 16000, "input": 1.5, "output": 3, "latency": 0.92, "throughput": 267.7 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.18, "output": 0.6, "latency": null, "throughput": null } ]	0.9	0.9	0.2	0.2	1.5	3	true
Infermatic: Mistral Nemo Inferor 12B	Infermatic/MN-12B-Inferor-v0.0	Infermatic	[ { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 2.69, "throughput": 14.62 } ]	0.8	1.2	0.8	1.2	0.8	1.2	true
Qwen2.5 Coder 32B Instruct	Qwen/Qwen2.5-Coder-32B-Instruct	Qwen	[ { "name": "Lambda", "context": 33000, "max_output": 3000, "input": 0.07, "output": 0.16, "latency": 0.5, "throughput": 75.8 }, { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.07, "output": 0.16, "latency": 0.26, "throughput": 50.78 }, { "name": "Hyperbolic", "context": 128000, "max_output": 8000, "input": 0.2, "output": 0.2, "latency": 1.14, "throughput": 54.07 }, { "name": "Parasail", "context": 128000, "max_output": 128000, "input": 0.5, "output": 0.5, "latency": 0.56, "throughput": 52.8 }, { "name": "Groq", "context": 131000, "max_output": 131000, "input": 0.79, "output": 0.79, "latency": 0.42, "throughput": 333.3 }, { "name": "Together", "context": 16000, "max_output": 2000, "input": 0.8, "output": 0.8, "latency": 0.5, "throughput": 80.49 }, { "name": "Fireworks", "context": 131000, "max_output": 4000, "input": 0.9, "output": 0.9, "latency": 0.45, "throughput": 55.73 }, { "name": "Mancer", "context": 33000, "max_output": 2000, "input": 1.5, "output": 2.813, "latency": 0.93, "throughput": 19.21 }, { "name": "SambaNova", "context": 16000, "max_output": 16000, "input": 1.5, "output": 3, "latency": 0.78, "throughput": 414.9 }, { "name": "Mancer (private)", "context": 33000, "max_output": 2000, "input": 2, "output": 3.75, "latency": null, "throughput": null }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 2.6, "output": 3.4, "latency": 2.06, "throughput": 13.11 } ]	0.8	0.8	0.07	0.16	2.6	3.4	true
SorcererLM 8x22B	rAIfle/SorcererLM-8x22b-bf16	rAIfle	[ { "name": "Infermatic", "context": 16000, "max_output": 16000, "input": 4.5, "output": 4.5, "latency": 2.83, "throughput": 8.29 } ]	4.5	4.5	4.5	4.5	4.5	4.5	true
EVA Qwen2.5 32B	EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2	EVA-UNIT-01	[ { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 2.6, "output": 3.4, "latency": 1.42, "throughput": 11.58 } ]	2.6	3.4	2.6	3.4	2.6	3.4	true
Unslopnemo 12B	TheDrummer/UnslopNemo-12B-v4.1	TheDrummer	[ { "name": "Infermatic", "context": 32000, "max_output": 32000, "input": 0.5, "output": 0.5, "latency": 0.51, "throughput": 80.43 } ]	0.5	0.5	0.5	0.5	0.5	0.5	true
NeverSleep: Lumimaid v0.2 70B	NeverSleep/Lumimaid-v0.2-70B	NeverSleep	[ { "name": "Mancer", "context": 16000, "max_output": 2000, "input": 3.375, "output": 4.5, "latency": 3.25, "throughput": 12.05 }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 4, "output": 6, "latency": 2.84, "throughput": 11.9 }, { "name": "Mancer (private)", "context": 16000, "max_output": 2000, "input": 4.5, "output": 6, "latency": 1.77, "throughput": 12.8 } ]	4	6	3.375	4.5	4.5	6	true
Magnum v4 72B	anthracite-org/magnum-v4-72b	anthracite-org	[ { "name": "Mancer", "context": 16000, "max_output": 1000, "input": 1.875, "output": 2.25, "latency": 0.87, "throughput": 13.91 }, { "name": "Mancer (private)", "context": 16000, "max_output": 1000, "input": 2.5, "output": 3, "latency": 0.89, "throughput": 13.84 }, { "name": "Infermatic", "context": 33000, "max_output": 33000, "input": 3, "output": 3, "latency": 0.77, "throughput": 22.54 }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 4, "output": 6, "latency": 2.81, "throughput": 11.74 } ]	3	3	1.875	2.25	4	6	true
Qwen2.5 7B Instruct	Qwen/Qwen2.5-7B-Instruct	Qwen	[ { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.025, "output": 0.05, "latency": 0.19, "throughput": 47.01 }, { "name": "Together", "context": 33000, "max_output": 2000, "input": 0.3, "output": 0.3, "latency": 0.4, "throughput": 123 } ]	0.3	0.3	0.025	0.05	0.3	0.3	true
NVIDIA: Llama 3.1 Nemotron 70B Instruct	nvidia/Llama-3.1-Nemotron-70B-Instruct-HF	nvidia	[ { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.12, "output": 0.3, "latency": 0.93, "throughput": 33.31 }, { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.12, "output": 0.3, "latency": 0.59, "throughput": 28.13 }, { "name": "Together", "context": 33000, "max_output": 33000, "input": 0.88, "output": 0.88, "latency": 0.54, "throughput": 70.39 }, { "name": "Infermatic", "context": 32000, "max_output": 32000, "input": 1, "output": 1, "latency": 2.63, "throughput": 13.9 } ]	0.88	0.88	0.12	0.3	1	1	true
Magnum v2 72B	anthracite-org/magnum-v2-72b	anthracite-org	[ { "name": "Infermatic", "context": 33000, "max_output": 33000, "input": 3, "output": 3, "latency": 2.98, "throughput": 42.32 }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 4, "output": 6, "latency": 6.41, "throughput": 12.61 } ]	4	6	3	3	4	6	true
Rocinante 12B	TheDrummer/Rocinante-12B-v1.1	TheDrummer	[ { "name": "Infermatic", "context": 33000, "max_output": 33000, "input": 0.25, "output": 0.5, "latency": 0.65, "throughput": 24.05 }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 2.45, "throughput": 11.45 } ]	0.8	1.2	0.25	0.5	0.8	1.2	true
Meta: Llama 3.2 3B Instruct	meta-llama/Llama-3.2-3B-Instruct	meta-llama	[ { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.015, "output": 0.025, "latency": 0.47, "throughput": 94.12 }, { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.015, "output": 0.025, "latency": 0.08, "throughput": 92.2 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.02, "output": 0.02, "latency": 0.87, "throughput": 85.37 }, { "name": "Lepton", "context": 131000, "max_output": 131000, "input": 0.03, "output": 0.03, "latency": 0.34, "throughput": 201 }, { "name": "Together", "context": 131000, "max_output": 16000, "input": 0.06, "output": 0.06, "latency": 1.08, "throughput": 58.03 }, { "name": "SambaNova", "context": 4000, "max_output": 2000, "input": 0.08, "output": 0.16, "latency": 0.25, "throughput": 1353 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.1, "output": 0.1, "latency": 0.55, "throughput": 156.5 }, { "name": "Hyperbolic", "context": 131000, "max_output": 131000, "input": 0.1, "output": 0.1, "latency": 1.26, "throughput": 187.4 }, { "name": "Cloudflare", "context": 131000, "max_output": 131000, "input": 0.1, "output": 0.1, "latency": 0.89, "throughput": 159.5 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.03, "output": 0.05, "latency": 1.26, "throughput": 79.98 } ]	0.06	0.06	0.015	0.025	0.08	0.16	true
Meta: Llama 3.2 1B Instruct	meta-llama/Llama-3.2-1B-Instruct	meta-llama	[ { "name": "Lepton", "context": 131000, "max_output": 131000, "input": 0.01, "output": 0.01, "latency": 0.33, "throughput": 421.3 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.01, "output": 0.01, "latency": 0.53, "throughput": 133.9 }, { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.01, "output": 0.01, "latency": 0.18, "throughput": 154.1 }, { "name": "SambaNova", "context": 4000, "max_output": 2000, "input": 0.04, "output": 0.08, "latency": 0.46, "throughput": 2093 }, { "name": "Cloudflare", "context": 131000, "max_output": 131000, "input": 0.1, "output": 0.1, "latency": 0.87, "throughput": 225.2 }, { "name": "NovitaAI", "context": 131000, "max_output": 131000, "input": 0.02, "output": 0.02, "latency": null, "throughput": null } ]	0.02	0.02	0.01	0.01	0.1	0.1	true
Meta: Llama 3.2 90B Vision Instruct	meta-llama/Llama-3.2-90B-Vision-Instruct	meta-llama	[ { "name": "SambaNova", "context": 4000, "max_output": 2000, "input": 0.8, "output": 1.6, "latency": 0.59, "throughput": 272.8 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.9, "output": 0.9, "latency": 0.86, "throughput": 34.05 }, { "name": "Together", "context": 131000, "max_output": 131000, "input": 1.2, "output": 1.2, "latency": 0.63, "throughput": 35.98 }, { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.35, "output": 0.4, "latency": 0.46, "throughput": 24.82 } ]	1.2	1.2	0.35	0.4	0.8	1.6	true
Meta: Llama 3.2 11B Vision Instruct	meta-llama/Llama-3.2-11B-Vision-Instruct	meta-llama	[ { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.055, "output": 0.055, "latency": 0.98, "throughput": 39.34 }, { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.055, "output": 0.055, "latency": 0.14, "throughput": 52.31 }, { "name": "SambaNova", "context": 4000, "max_output": 2000, "input": 0.15, "output": 0.3, "latency": 0.62, "throughput": 869.8 }, { "name": "Together", "context": 131000, "max_output": 131000, "input": 0.18, "output": 0.18, "latency": 0.34, "throughput": 162.1 }, { "name": "Cloudflare", "context": 131000, "max_output": 131000, "input": 0.2, "output": 0.2, "latency": 1.49, "throughput": 35.49 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.2, "output": 0.2, "latency": 5.15, "throughput": 35.08 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.06, "output": 0.06, "latency": null, "throughput": null } ]	0.18	0.18	0.055	0.055	0.15	0.3	true
Qwen2.5 72B Instruct	Qwen/Qwen2.5-72B-Instruct	Qwen	[ { "name": "Nebius AI Studio", "context": 128000, "max_output": 128000, "input": 0.13, "output": 0.4, "latency": 0.21, "throughput": 29.47 }, { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.13, "output": 0.4, "latency": 0.44, "throughput": 36.79 }, { "name": "Hyperbolic", "context": 33000, "max_output": 33000, "input": 0.4, "output": 0.4, "latency": 1.34, "throughput": 44.11 }, { "name": "Fireworks", "context": 33000, "max_output": 33000, "input": 0.9, "output": 0.9, "latency": 0.39, "throughput": 38.62 }, { "name": "Together", "context": 131000, "max_output": 2000, "input": 1.2, "output": 1.2, "latency": 0.43, "throughput": 96.85 }, { "name": "SambaNova", "context": 16000, "max_output": 16000, "input": 2, "output": 4, "latency": 0.41, "throughput": 359.3 }, { "name": "NovitaAI", "context": 32000, "max_output": 4000, "input": 0.38, "output": 0.4, "latency": 0.83, "throughput": 20.55 } ]	0.4	0.4	0.13	0.4	2	4	true
Qwen: Qwen2-VL 72B Instruct	Qwen/Qwen2-VL-72B-Instruct	Qwen	[ { "name": "Hyperbolic", "context": 4000, "max_output": 4000, "input": 0.4, "output": 0.4, "latency": 1.21, "throughput": 37.83 } ]	0.4	0.4	0.4	0.4	0.4	0.4	true
NeverSleep: Lumimaid v0.2 8B	NeverSleep/Lumimaid-v0.2-8B	NeverSleep	[ { "name": "Mancer", "context": 33000, "max_output": 2000, "input": 0.1875, "output": 1.125, "latency": 0.95, "throughput": 36.34 }, { "name": "Mancer (private)", "context": 33000, "max_output": 2000, "input": 0.25, "output": 1.5, "latency": 0.69, "throughput": 36.51 }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 1.6, "throughput": 26.34 } ]	0.25	1.5	0.1875	1.125	0.8	1.2	true
Mistral: Pixtral 12B	mistralai/Pixtral-12B-2409	mistralai	[ { "name": "Hyperbolic", "context": 4000, "max_output": 4000, "input": 0.1, "output": 0.1, "latency": 1.4, "throughput": 52.28 }, { "name": "Mistral", "context": 4000, "max_output": 4000, "input": 0.15, "output": 0.15, "latency": 1.13, "throughput": 90.11 } ]	0.15	0.15	0.1	0.1	0.15	0.15	true
Sao10K: Llama 3.1 Euryale 70B v2.2	Sao10K/L3.1-70B-Euryale-v2.2	Sao10K	[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.7, "output": 0.8, "latency": 0.32, "throughput": 37.38 }, { "name": "Infermatic", "context": 16000, "max_output": 16000, "input": 1.5, "output": 1.5, "latency": 0.93, "throughput": 19.76 }, { "name": "NovitaAI", "context": 16000, "max_output": 16000, "input": 1.48, "output": 1.48, "latency": 1.51, "throughput": 20.91 } ]	1.48	1.48	0.7	0.8	1.5	1.5	true
Qwen: Qwen2-VL 7B Instruct	Qwen/Qwen2-VL-7B-Instruct	Qwen	[ { "name": "Hyperbolic", "context": 4000, "max_output": 4000, "input": 0.1, "output": 0.1, "latency": 1, "throughput": 93.31 } ]	0.1	0.1	0.1	0.1	0.1	0.1	true
Nous: Hermes 3 70B Instruct	NousResearch/Hermes-3-Llama-3.1-70B	NousResearch	[ { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.12, "output": 0.3, "latency": 0.74, "throughput": 32.01 }, { "name": "Hyperbolic", "context": 12000, "max_output": 12000, "input": 0.4, "output": 0.4, "latency": 0.95, "throughput": 31.07 } ]	0.4	0.4	0.12	0.3	0.4	0.4	true
Nous: Hermes 3 405B Instruct	NousResearch/Hermes-3-Llama-3.1-405B	NousResearch	[ { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.8, "output": 0.8, "latency": 1.1, "throughput": 26.23 }, { "name": "inference.net", "context": 33000, "max_output": 33000, "input": 0.8, "output": 0.8, "latency": 0.92, "throughput": 25.74 }, { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.8, "output": 0.8, "latency": 1.01, "throughput": 11.46 }, { "name": "Nebius AI Studio", "context": 131000, "max_output": 131000, "input": 1, "output": 3, "latency": 0.31, "throughput": 26.61 } ]	0.8	0.8	0.8	0.8	1	3	true
Sao10K: Llama 3 8B Lunaris	Sao10K/L3-8B-Lunaris-v1	Sao10K	[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.03, "output": 0.06, "latency": 0.38, "throughput": 73.74 }, { "name": "NovitaAI", "context": 8000, "max_output": 8000, "input": 0.05, "output": 0.05, "latency": 0.88, "throughput": 59.26 } ]	0.05	0.05	0.03	0.06	0.05	0.05	true
Aetherwiing: Starcannon 12B	intervitens/mini-magnum-12b-v1.1	intervitens	[ { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 3.92, "throughput": 15.83 } ]	0.8	1.2	0.8	1.2	0.8	1.2	true
Meta: Llama 3.1 405B (base)	meta-llama/llama-3.1-405B	meta-llama	[ { "name": "Hyperbolic (quantized)", "context": 33000, "max_output": 33000, "input": 2, "output": 2, "latency": 0.74, "throughput": 23.31 }, { "name": "Hyperbolic", "context": 33000, "max_output": 33000, "input": 4, "output": 4, "latency": 1.19, "throughput": 16.15 } ]	4	4	2	2	4	4	true
Mistral Nemo 12B Celeste	nothingiisreal/MN-12B-Celeste-V1.9	nothingiisreal	[ { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 2.54, "throughput": 14.35 } ]	0.8	1.2	0.8	1.2	0.8	1.2	true
Meta: Llama 3.1 405B Instruct	meta-llama/Meta-Llama-3.1-405B-Instruct	meta-llama	[ { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.8, "output": 0.8, "latency": 0.15, "throughput": 17.38 }, { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.8, "output": 0.8, "latency": 0.64, "throughput": 34.89 }, { "name": "Nebius AI Studio", "context": 128000, "max_output": 128000, "input": 1, "output": 3, "latency": 0.15, "throughput": 27.72 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 3, "output": 3, "latency": 1.12, "throughput": 44.58 }, { "name": "Together", "context": 131000, "max_output": 131000, "input": 3.5, "output": 3.5, "latency": 1.08, "throughput": 49.02 }, { "name": "kluster.ai", "context": 131000, "max_output": 131000, "input": 3.5, "output": 3.5, "latency": 1.41, "throughput": 17.25 }, { "name": "Hyperbolic", "context": 33000, "max_output": 33000, "input": 4, "output": 4, "latency": 1.02, "throughput": 8.78 }, { "name": "SambaNova", "context": 8000, "max_output": 4000, "input": 5, "output": 10, "latency": 1.53, "throughput": 81.81 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 2.75, "output": 2.75, "latency": null, "throughput": null } ]	3	3	0.8	0.8	5	10	true
Meta: Llama 3.1 8B Instruct	meta-llama/Meta-Llama-3.1-8B-Instruct	meta-llama	[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.02, "output": 0.05, "latency": 0.08, "throughput": 74.22 }, { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.025, "output": 0.04, "latency": 0.43, "throughput": 220.6 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.03, "output": 0.03, "latency": 0.63, "throughput": 59.25 }, { "name": "Groq", "context": 131000, "max_output": 8000, "input": 0.05, "output": 0.08, "latency": 0.3, "throughput": 552.2 }, { "name": "Lepton", "context": 131000, "max_output": 131000, "input": 0.07, "output": 0.07, "latency": 0.21, "throughput": 100.1 }, { "name": "Friendli", "context": 131000, "max_output": 131000, "input": 0.1, "output": 0.1, "latency": 0.42, "throughput": 245.1 }, { "name": "Hyperbolic", "context": 33000, "max_output": 33000, "input": 0.1, "output": 0.1, "latency": 1.19, "throughput": 101 }, { "name": "SambaNova", "context": 8000, "max_output": 4000, "input": 0.1, "output": 0.2, "latency": 0.28, "throughput": 803.3 }, { "name": "Cloudflare", "context": 131000, "max_output": 131000, "input": 0.15, "output": 0.15, "latency": 0.31, "throughput": 126.3 }, { "name": "kluster.ai", "context": 131000, "max_output": 131000, "input": 0.18, "output": 0.18, "latency": 0.32, "throughput": 13.41 }, { "name": "Together", "context": 131000, "max_output": 131000, "input": 0.18, "output": 0.18, "latency": 0.25, "throughput": 313.2 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.2, "output": 0.2, "latency": 0.36, "throughput": 223.2 }, { "name": "Avian.io", "context": 131000, "max_output": 131000, "input": 0.2, "output": 0.2, "latency": 0.29, "throughput": 308 }, { "name": "NovitaAI", "context": 16000, "max_output": 8000, "input": 0.05, "output": 0.05, "latency": 1.84, "throughput": 66.66 } ]	0.1	0.1	0.03	0.03	0.2	0.2	true
Meta: Llama 3.1 70B Instruct	meta-llama/Meta-Llama-3.1-70B-Instruct	meta-llama	[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.12, "output": 0.3, "latency": 0.54, "throughput": 33.04 }, { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.12, "output": 0.3, "latency": 0.58, "throughput": 37.2 }, { "name": "Nebius AI Studio", "context": 128000, "max_output": 128000, "input": 0.13, "output": 0.4, "latency": 0.15, "throughput": 29.24 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.4, "output": 0.4, "latency": 1.38, "throughput": 16.81 }, { "name": "Hyperbolic", "context": 33000, "max_output": 33000, "input": 0.4, "output": 0.4, "latency": 0.88, "throughput": 106.8 }, { "name": "Friendli", "context": 131000, "max_output": 131000, "input": 0.6, "output": 0.6, "latency": 0.2, "throughput": 113.5 }, { "name": "SambaNova", "context": 8000, "max_output": 4000, "input": 0.6, "output": 1.2, "latency": 0.53, "throughput": 269.4 }, { "name": "Cloudflare", "context": 131000, "max_output": 131000, "input": 0.75, "output": 0.75, "latency": 0.78, "throughput": 27.26 }, { "name": "Lepton", "context": 131000, "max_output": 131000, "input": 0.8, "output": 0.8, "latency": 0.3, "throughput": 45.24 }, { "name": "Together", "context": 131000, "max_output": 131000, "input": 0.88, "output": 0.88, "latency": 0.33, "throughput": 115.5 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.9, "output": 0.9, "latency": 0.44, "throughput": 94.56 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.34, "output": 0.39, "latency": 1.72, "throughput": 52.16 } ]	0.6	0.6	0.12	0.3	0.9	0.9	true
Mistral: Mistral Nemo	mistralai/Mistral-Nemo-Instruct-2407	mistralai	[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.035, "output": 0.08, "latency": 0.25, "throughput": 52.49 }, { "name": "Nebius AI Studio", "context": 128000, "max_output": 128000, "input": 0.04, "output": 0.12, "latency": 0.64, "throughput": 22.83 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.1, "output": 0.1, "latency": 0.69, "throughput": 63.6 }, { "name": "Parasail", "context": 128000, "max_output": 128000, "input": 0.11, "output": 0.11, "latency": 0.75, "throughput": 134.1 }, { "name": "Mistral", "context": 128000, "max_output": 128000, "input": 0.15, "output": 0.15, "latency": 0.29, "throughput": 128.1 }, { "name": "Lepton", "context": 128000, "max_output": 128000, "input": 0.18, "output": 0.18, "latency": 0.28, "throughput": 50.18 }, { "name": "Azure", "context": 128000, "max_output": 128000, "input": 0.3, "output": 0.3, "latency": 0.64, "throughput": 90.2 }, { "name": "NovitaAI", "context": 131000, "max_output": 131000, "input": 0.17, "output": 0.17, "latency": 1.61, "throughput": 39.52 } ]	0.15	0.15	0.035	0.08	0.3	0.3	true
Mistral: Codestral Mamba	mistralai/mamba-codestral-7B-v0.1	mistralai	[ { "name": "Mistral", "context": 256000, "max_output": 256000, "input": 0.25, "output": 0.25, "latency": 0.34, "throughput": 97.73 } ]	0.25	0.25	0.25	0.25	0.25	0.25	true
Qwen 2 7B Instruct	Qwen/Qwen2-7B-Instruct	Qwen	[ { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.054, "output": 0.054, "latency": 0.65, "throughput": 82.28 } ]	0.054	0.054	0.054	0.054	0.054	0.054	true
Google: Gemma 2 27B	google/gemma-2-27b-it	google	[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.27, "output": 0.27, "latency": 0.49, "throughput": 38.62 }, { "name": "Together", "context": 8000, "max_output": 2000, "input": 0.8, "output": 0.8, "latency": 0.41, "throughput": 69.52 } ]	0.8	0.8	0.27	0.27	0.8	0.8	true
Magnum 72B	alpindale/magnum-72b-v1	alpindale	[ { "name": "Mancer", "context": 16000, "max_output": 1000, "input": 1.875, "output": 2.25, "latency": 1.21, "throughput": 22.84 }, { "name": "Mancer (private)", "context": 16000, "max_output": 1000, "input": 2.5, "output": 3, "latency": 0.92, "throughput": 22.6 }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 4, "output": 6, "latency": 8.11, "throughput": 11.31 } ]	2.5	3	1.875	2.25	4	6	true
Google: Gemma 2 9B	google/gemma-2-9b-it	google	[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.03, "output": 0.06, "latency": 0.09, "throughput": 35.94 }, { "name": "Lepton", "context": 8000, "max_output": 8000, "input": 0.07, "output": 0.07, "latency": 0.27, "throughput": 103.4 }, { "name": "Groq", "context": 8000, "max_output": 8000, "input": 0.2, "output": 0.2, "latency": 0.22, "throughput": 560.1 }, { "name": "Together", "context": 8000, "max_output": 8000, "input": 0.3, "output": 0.3, "latency": 0.22, "throughput": 109 }, { "name": "NovitaAI", "context": 8000, "max_output": 8000, "input": 0.08, "output": 0.08, "latency": 4.21, "throughput": 29.98 } ]	0.08	0.08	0.03	0.06	0.3	0.3	true
Sao10k: Llama 3 Euryale 70B v2.1	Sao10K/L3-70B-Euryale-v2.1	Sao10K	[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.7, "output": 0.8, "latency": 0.2, "throughput": 40.32 }, { "name": "NovitaAI", "context": 16000, "max_output": 16000, "input": 1.48, "output": 1.48, "latency": 3.58, "throughput": 19.49 } ]	1.48	1.48	0.7	0.8	1.48	1.48	true
Dolphin 2.9.2 Mixtral 8x22B 🐬	cognitivecomputations/dolphin-2.9.2-mixtral-8x22b	cognitivecomputations	[ { "name": "NovitaAI", "context": 16000, "max_output": 16000, "input": 0.9, "output": 0.9, "latency": 6.55, "throughput": 9.32 } ]	0.9	0.9	0.9	0.9	0.9	0.9	true
Qwen 2 72B Instruct	Qwen/Qwen2-72B-Instruct	Qwen	[ { "name": "Together", "context": 33000, "max_output": 4000, "input": 0.9, "output": 0.9, "latency": 0.45, "throughput": 67.28 } ]	0.9	0.9	0.9	0.9	0.9	0.9	true
Mistral: Mistral 7B Instruct	mistralai/Mistral-7B-Instruct-v0.3	mistralai	[ { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.03, "output": 0.055, "latency": 0.29, "throughput": 82.05 }, { "name": "Parasail", "context": 33000, "max_output": 33000, "input": 0.11, "output": 0.11, "latency": 0.63, "throughput": 128.1 }, { "name": "Together", "context": 33000, "max_output": 4000, "input": 0.2, "output": 0.2, "latency": 0.37, "throughput": 163.2 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.059, "output": 0.059, "latency": 1.1, "throughput": 121.5 }, { "name": "Lepton", "context": 33000, "max_output": 33000, "input": 0.07, "output": 0.07, "latency": 0.54, "throughput": 105.4 } ]	0.07	0.07	0.03	0.055	0.2	0.2	true
Mistral: Mistral 7B Instruct v0.3	mistralai/Mistral-7B-Instruct-v0.3	mistralai	[ { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.03, "output": 0.055, "latency": 0.2, "throughput": 91.8 }, { "name": "Lepton", "context": 33000, "max_output": 33000, "input": 0.07, "output": 0.07, "latency": 0.28, "throughput": 107.9 }, { "name": "Together", "context": 33000, "max_output": 4000, "input": 0.2, "output": 0.2, "latency": 0.46, "throughput": 136.8 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.059, "output": 0.059, "latency": null, "throughput": null } ]	0.07	0.07	0.03	0.055	0.2	0.2	true
NousResearch: Hermes 2 Pro - Llama-3 8B	NousResearch/Hermes-2-Pro-Llama-3-8B	NousResearch	[ { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.025, "output": 0.04, "latency": 0.5, "throughput": 271.2 }, { "name": "NovitaAI", "context": 8000, "max_output": 8000, "input": 0.14, "output": 0.14, "latency": 1.23, "throughput": 126.7 } ]	0.14	0.14	0.025	0.04	0.14	0.14	true
Microsoft: Phi-3 Mini 128K Instruct	microsoft/Phi-3-mini-128k-instruct	microsoft	[ { "name": "Azure", "context": 128000, "max_output": 128000, "input": 0.1, "output": 0.1, "latency": 0.48, "throughput": 91.92 } ]	0.1	0.1	0.1	0.1	0.1	0.1	true
Microsoft: Phi-3 Medium 128K Instruct	microsoft/Phi-3-medium-128k-instruct	microsoft	[ { "name": "Azure", "context": 128000, "max_output": 128000, "input": 1, "output": 1, "latency": 0.56, "throughput": 58.03 } ]	1	1	1	1	1	1	true
NeverSleep: Llama 3 Lumimaid 70B	NeverSleep/Llama-3-Lumimaid-70B-v0.1	NeverSleep	[ { "name": "Mancer", "context": 8000, "max_output": 2000, "input": 3.375, "output": 4.5, "latency": 0.91, "throughput": 13.09 }, { "name": "Featherless", "context": 8000, "max_output": 4000, "input": 4, "output": 6, "latency": 2.25, "throughput": 9.73 }, { "name": "Mancer (private)", "context": 8000, "max_output": 2000, "input": 4.5, "output": 6, "latency": 0.82, "throughput": 13.63 } ]	4	6	3.375	4.5	4.5	6	true
Meta: LlamaGuard 2 8B	meta-llama/LlamaGuard-7b	meta-llama	[ { "name": "Together", "context": 8000, "max_output": 8000, "input": 0.2, "output": 0.2, "latency": 0.98, "throughput": 76.25 } ]	0.2	0.2	0.2	0.2	0.2	0.2	true
NeverSleep: Llama 3 Lumimaid 8B (extended)	NeverSleep/Llama-3-Lumimaid-8B-v0.1	NeverSleep	[ { "name": "Mancer", "context": 25000, "max_output": 2000, "input": 0.1875, "output": 1.125, "latency": 0.77, "throughput": 50.08 }, { "name": "Mancer (private)", "context": 25000, "max_output": 2000, "input": 0.25, "output": 1.5, "latency": 0.54, "throughput": 53.36 } ]	0.25	1.5	0.1875	1.125	0.25	1.5	true
NeverSleep: Llama 3 Lumimaid 8B	NeverSleep/Llama-3-Lumimaid-8B-v0.1	NeverSleep	[ { "name": "Mancer", "context": 25000, "max_output": 2000, "input": 0.1875, "output": 1.125, "latency": 0.76, "throughput": 50.01 }, { "name": "Mancer (private)", "context": 25000, "max_output": 2000, "input": 0.25, "output": 1.5, "latency": 0.54, "throughput": 53.36 }, { "name": "Featherless", "context": 8000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 1.3, "throughput": 25.6 } ]	0.25	1.5	0.1875	1.125	0.8	1.2	true
Fimbulvetr 11B v2	Sao10K/Fimbulvetr-11B-v2	Sao10K	[ { "name": "Featherless", "context": 4000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 1.74, "throughput": 24.72 } ]	0.8	1.2	0.8	1.2	0.8	1.2	true
Meta: Llama 3 8B Instruct	meta-llama/Meta-Llama-3-8B-Instruct	meta-llama	[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.03, "output": 0.06, "latency": 0.17, "throughput": 115.6 }, { "name": "Groq", "context": 8000, "max_output": 8000, "input": 0.05, "output": 0.08, "latency": 0.29, "throughput": 478.4 }, { "name": "Cloudflare", "context": 8000, "max_output": 8000, "input": 0.15, "output": 0.15, "latency": 0.72, "throughput": 15.63 }, { "name": "Mancer", "context": 16000, "max_output": 2000, "input": 0.1875, "output": 1.125, "latency": 0.56, "throughput": 51.09 }, { "name": "Fireworks", "context": 8000, "max_output": 8000, "input": 0.2, "output": 0.2, "latency": 0.48, "throughput": 146.5 }, { "name": "Mancer (private)", "context": 16000, "max_output": 2000, "input": 0.25, "output": 1.5, "latency": 0.6, "throughput": 51.02 }, { "name": "NovitaAI", "context": 8000, "max_output": 8000, "input": 0.04, "output": 0.04, "latency": null, "throughput": null } ]	0.15	0.15	0.04	0.04	0.25	1.5	true
Meta: Llama 3 70B Instruct	meta-llama/Meta-Llama-3-70B-Instruct	meta-llama	[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.23, "output": 0.4, "latency": 0.41, "throughput": 24.94 }, { "name": "Groq", "context": 8000, "max_output": 8000, "input": 0.59, "output": 0.79, "latency": 0.22, "throughput": 265 }, { "name": "Together", "context": 8000, "max_output": 8000, "input": 0.88, "output": 0.88, "latency": 0.42, "throughput": 20.15 }, { "name": "Fireworks", "context": 8000, "max_output": 8000, "input": 0.9, "output": 0.9, "latency": 0.59, "throughput": 162.2 }, { "name": "NovitaAI", "context": 8000, "max_output": 8000, "input": 0.51, "output": 0.74, "latency": 4.84, "throughput": 20.38 } ]	0.59	0.79	0.23	0.4	0.9	0.9	true
Mistral: Mixtral 8x22B Instruct	mistralai/Mixtral-8x22B-Instruct-v0.1	mistralai	[ { "name": "Fireworks", "context": 66000, "max_output": 66000, "input": 0.9, "output": 0.9, "latency": 0.67, "throughput": 77.48 }, { "name": "Together", "context": 66000, "max_output": 2000, "input": 1.2, "output": 1.2, "latency": 0.78, "throughput": 104.2 }, { "name": "Mistral", "context": 66000, "max_output": 66000, "input": 2, "output": 6, "latency": 0.26, "throughput": 97.77 } ]	1.2	1.2	0.9	0.9	2	6	true
WizardLM-2 8x22B	microsoft/WizardLM-2-8x22B	microsoft	[ { "name": "DeepInfra", "context": 66000, "max_output": 8000, "input": 0.5, "output": 0.5, "latency": 0.1, "throughput": 29.57 }, { "name": "Lepton", "context": 66000, "max_output": 66000, "input": 1, "output": 1, "latency": 0.39, "throughput": 33.75 }, { "name": "NovitaAI", "context": 66000, "max_output": 66000, "input": 0.62, "output": 0.62, "latency": 1.06, "throughput": 17.49 }, { "name": "Together", "context": 66000, "max_output": 66000, "input": 1.2, "output": 1.2, "latency": 0.56, "throughput": 64.35 } ]	1	1	0.5	0.5	1.2	1.2	true
WizardLM-2 7B	microsoft/WizardLM-2-7B	microsoft	[ { "name": "Lepton", "context": 32000, "max_output": 32000, "input": 0.07, "output": 0.07, "latency": 0.36, "throughput": 102.4 } ]	0.07	0.07	0.07	0.07	0.07	0.07	true
Databricks: DBRX 132B Instruct	databricks/dbrx-instruct	databricks	[ { "name": "Together", "context": 33000, "max_output": 2000, "input": 1.2, "output": 1.2, "latency": 0.42, "throughput": 83.57 } ]	1.2	1.2	1.2	1.2	1.2	1.2	true
Midnight Rose 70B	sophosympatheia/Wizard-Tulu-Dolphin-70B-v1.0	sophosympatheia	[ { "name": "NovitaAI", "context": 4000, "max_output": 4000, "input": 0.8, "output": 0.8, "latency": 1.84, "throughput": 14.85 } ]	0.8	0.8	0.8	0.8	0.8	0.8	true
Google: Gemma 7B	google/gemma-1.1-7b-it	google	[ { "name": "Cloudflare", "context": 8000, "max_output": 8000, "input": 0.15, "output": 0.15, "latency": 1.33, "throughput": 13.07 } ]	0.15	0.15	0.15	0.15	0.15	0.15	true
Nous: Hermes 2 Mixtral 8x7B DPO	NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO	NousResearch	[ { "name": "Together", "context": 33000, "max_output": 2000, "input": 0.6, "output": 0.6, "latency": 0.45, "throughput": 105.8 } ]	0.6	0.6	0.6	0.6	0.6	0.6	true
Dolphin 2.6 Mixtral 8x7B 🐬	cognitivecomputations/dolphin-2.6-mixtral-8x7b	cognitivecomputations	[ { "name": "Lepton", "context": 33000, "max_output": 33000, "input": 0.5, "output": 0.5, "latency": 0.11, "throughput": 84.13 } ]	0.5	0.5	0.5	0.5	0.5	0.5	true
Mistral: Mixtral 8x7B (base)	mistralai/Mixtral-8x7B-v0.1	mistralai	[ { "name": "Together", "context": 33000, "max_output": 2000, "input": 0.6, "output": 0.6, "latency": 0.38, "throughput": 128 } ]	0.6	0.6	0.6	0.6	0.6	0.6	true
Mistral: Mixtral 8x7B Instruct	mistralai/Mixtral-8x7B-Instruct-v0.1	mistralai	[ { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.24, "output": 0.24, "latency": 0.31, "throughput": 101.4 }, { "name": "Groq", "context": 33000, "max_output": 33000, "input": 0.24, "output": 0.24, "latency": 0.34, "throughput": 676.8 }, { "name": "Fireworks", "context": 33000, "max_output": 33000, "input": 0.5, "output": 0.5, "latency": 0.62, "throughput": 155.2 }, { "name": "Lepton", "context": 33000, "max_output": 33000, "input": 0.5, "output": 0.5, "latency": 0.31, "throughput": 73.77 }, { "name": "Together", "context": 33000, "max_output": 2000, "input": 0.6, "output": 0.6, "latency": 0.51, "throughput": 92.37 } ]	0.5	0.5	0.24	0.24	0.6	0.6	true
OpenChat 3.5 7B	openchat/openchat-3.5-0106	openchat	[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.055, "output": 0.055, "latency": 0.23, "throughput": 98.01 }, { "name": "Lepton", "context": 8000, "max_output": 8000, "input": 0.07, "output": 0.07, "latency": 0.41, "throughput": 106.5 }, { "name": "NovitaAI", "context": 4000, "max_output": 4000, "input": 0.06, "output": 0.06, "latency": 0.8, "throughput": 58.78 } ]	0.06	0.06	0.055	0.055	0.07	0.07	true
Noromaid 20B	NeverSleep/Noromaid-20b-v0.1.1	NeverSleep	[ { "name": "Mancer", "context": 8000, "max_output": 2000, "input": 1.5, "output": 2.25, "latency": 0.55, "throughput": 26.36 }, { "name": "Mancer (private)", "context": 8000, "max_output": 2000, "input": 2, "output": 3, "latency": 1.05, "throughput": 25.55 } ]	2	3	1.5	2.25	2	3	true
OpenHermes 2.5 Mistral 7B	teknium/OpenHermes-2.5-Mistral-7B	teknium	[ { "name": "NovitaAI", "context": 4000, "max_output": 4000, "input": 0.17, "output": 0.17, "latency": 0.66, "throughput": 148.9 } ]	0.17	0.17	0.17	0.17	0.17	0.17	true
Toppy M 7B	Undi95/Toppy-M-7B	Undi95	[ { "name": "Lepton", "context": 4000, "max_output": 4000, "input": 0.07, "output": 0.07, "latency": 0.41, "throughput": 107.1 } ]	0.07	0.07	0.07	0.07	0.07	0.07	true
Goliath 120B	alpindale/goliath-120b	alpindale	[ { "name": "Mancer", "context": 6000, "max_output": 512, "input": 9.375, "output": 9.375, "latency": 1.44, "throughput": 17.83 }, { "name": "Mancer (private)", "context": 6000, "max_output": 512, "input": 12.5, "output": 12.5, "latency": 1.29, "throughput": 15.68 } ]	12.5	12.5	9.375	9.375	12.5	12.5	true
Airoboros 70B	jondurbin/airoboros-l2-70b-2.2.1	jondurbin	[ { "name": "NovitaAI", "context": 4000, "max_output": 4000, "input": 0.5, "output": 0.5, "latency": 2.14, "throughput": 56.12 } ]	0.5	0.5	0.5	0.5	0.5	0.5	true
Xwin 70B	Xwin-LM/Xwin-LM-70B-V0.1	Xwin-LM	[ { "name": "Mancer", "context": 8000, "max_output": 512, "input": 3.75, "output": 3.75, "latency": 2.52, "throughput": 18.03 }, { "name": "Mancer (private)", "context": 8000, "max_output": 512, "input": 5, "output": 5, "latency": 2.09, "throughput": 18.19 } ]	5	5	3.75	3.75	5	5	true
Mistral: Mistral 7B Instruct v0.1	mistralai/Mistral-7B-Instruct-v0.1	mistralai	[ { "name": "Together", "context": 33000, "max_output": 2000, "input": 0.2, "output": 0.2, "latency": 0.31, "throughput": 178.2 } ]	0.2	0.2	0.2	0.2	0.2	0.2	true
Pygmalion: Mythalion 13B	PygmalionAI/mythalion-13b	PygmalionAI	[ { "name": "Featherless", "context": 4000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 1.86, "throughput": 12.32 }, { "name": "Mancer", "context": 8000, "max_output": 512, "input": 1.125, "output": 1.125, "latency": 0.92, "throughput": 22.86 }, { "name": "Mancer (private)", "context": 8000, "max_output": 512, "input": 1.5, "output": 1.5, "latency": 0.74, "throughput": 22.35 } ]	1.125	1.125	0.8	1.2	1.5	1.5	true
Nous: Hermes 13B	NousResearch/Nous-Hermes-Llama2-13b	NousResearch	[ { "name": "Lepton", "context": 4000, "max_output": 4000, "input": 0.18, "output": 0.18, "latency": 0.26, "throughput": 81.6 }, { "name": "NovitaAI", "context": 4000, "max_output": 4000, "input": 0.17, "output": 0.17, "latency": 0.83, "throughput": 82.68 } ]	0.18	0.18	0.17	0.17	0.18	0.18	true
ReMM SLERP 13B	Undi95/ReMM-SLERP-L2-13B	Undi95	[ { "name": "Featherless", "context": 4000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": null, "throughput": null }, { "name": "Mancer", "context": 6000, "max_output": 512, "input": 1.125, "output": 1.125, "latency": 0.6, "throughput": 43.76 }, { "name": "Mancer (private)", "context": 6000, "max_output": 512, "input": 1.5, "output": 1.5, "latency": 0.57, "throughput": 43.3 } ]	1.125	1.125	0.8	1.2	1.5	1.5	true
MythoMax 13B	Gryphe/MythoMax-L2-13b	Gryphe	[ { "name": "DeepInfra", "context": 4000, "max_output": 4000, "input": 0.065, "output": 0.065, "latency": 0.2, "throughput": 108.8 }, { "name": "Parasail", "context": 4000, "max_output": 4000, "input": 0.11, "output": 0.11, "latency": 0.56, "throughput": 91.29 }, { "name": "Lepton", "context": 4000, "max_output": 4000, "input": 0.18, "output": 0.18, "latency": 0.21, "throughput": 104.1 }, { "name": "Fireworks", "context": 4000, "max_output": 4000, "input": 0.2, "output": 0.2, "latency": 2.02, "throughput": 7.59 }, { "name": "Together", "context": 4000, "max_output": 4000, "input": 0.3, "output": 0.3, "latency": 0.45, "throughput": 123.7 }, { "name": "Mancer", "context": 8000, "max_output": 512, "input": 1.125, "output": 1.125, "latency": 0.59, "throughput": 41.95 }, { "name": "Mancer (private)", "context": 8000, "max_output": 512, "input": 1.5, "output": 1.5, "latency": 0.92, "throughput": 41.69 }, { "name": "NovitaAI", "context": 4000, "max_output": 4000, "input": 0.09, "output": 0.09, "latency": 1.9, "throughput": 79.17 } ]	0.2	0.2	0.065	0.065	1.5	1.5	true
Meta: Llama 2 13B Chat	meta-llama/Llama-2-13b-chat-hf	meta-llama	[ { "name": "Together", "context": 4000, "max_output": 2000, "input": 0.22, "output": 0.22, "latency": 0.76, "throughput": 63.12 }, { "name": "Lepton", "context": 4000, "max_output": 4000, "input": 0.3, "output": 0.3, "latency": 0.44, "throughput": 98.15 } ]	0.3	0.3	0.22	0.22	0.3	0.3	true
Meta: Llama 2 70B Chat	meta-llama/Llama-2-70b-chat-hf	meta-llama	[ { "name": "Together", "context": 4000, "max_output": 4000, "input": 0.9, "output": 0.9, "latency": 0.73, "throughput": 45.19 } ]	0.9	0.9	0.9	0.9	0.9	0.9	true
Perplexity: Sonar Reasoning Pro	null	perplexity-ai	[ { "name": "Perplexity", "context": 128000, "max_output": 128000, "input": 2, "output": 8, "latency": 11.05, "throughput": 91.82 } ]	2	8	2	8	2	8	false
Perplexity: Sonar Pro	null	perplexity-ai	[ { "name": "Perplexity", "context": 200000, "max_output": 8000, "input": 3, "output": 15, "latency": 2.49, "throughput": 51.06 } ]	3	15	3	15	3	15	false
Perplexity: Sonar Deep Research	null	perplexity-ai	[ { "name": "Perplexity", "context": 200000, "max_output": 200000, "input": 2, "output": 8, "latency": 24.87, "throughput": 51.64 } ]	2	8	2	8	2	8	false
OpenAI: GPT-4.5 (Preview)	null	OpenAI	[ { "name": "OpenAI", "context": 128000, "max_output": 16000, "input": 75, "output": 150, "latency": 1.64, "throughput": 14.34 } ]	75	150	75	150	75	150	false

Qwen: QwQ 32B

Qwen/QwQ-32B

Qwen

[ { "name": "Nebius AI Studio", "context": 131000, "max_output": 131000, "input": 0.15, "output": 0.45, "latency": 0.38, "throughput": 40.22 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.18, "output": 0.2, "latency": 1.4, "throughput": 20.63 }, { "name": "Groq", "context": 131000, "max_output": 131000, "input": 0.29, "output": 0.39, "latency": 0.38, "throughput": 438.3 }, { "name": "Hyperbolic", "context": 131000, "max_output": 131000, "input": 0.4, "output": 0.4, "latency": 0.73, "throughput": 26.72 }, { "name": "Parasail", "context": 131000, "max_output": 131000, "input": 0.5, "output": 0.5, "latency": 3.22, "throughput": 32.44 }, { "name": "Nebius AI Studio (Fast)", "context": 131000, "max_output": 131000, "input": 0.5, "output": 1.5, "latency": 0.32, "throughput": 83.05 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.9, "output": 0.9, "latency": 0.82, "throughput": 124.8 } ]

0.4

0.18

0.2

0.5

1.5

true

Perplexity: R1 1776

perplexity-ai/r1-1776

perplexity-ai

[ { "name": "Perplexity", "context": 128000, "max_output": 128000, "input": 2, "output": 8, "latency": 6.35, "throughput": 85.07 } ]

2

8

2

8

2

8

true

Llama Guard 3 8B

meta-llama/Llama-Guard-3-8B

meta-llama

[ { "name": "Groq", "context": 8000, "max_output": 8000, "input": 0.2, "output": 0.2, "latency": null, "throughput": null }, { "name": "SambaNova", "context": 16000, "max_output": 16000, "input": 0.3, "output": 0.3, "latency": 0.57, "throughput": 897.3 } ]

0.3

0.2

0.3

true

Llama 3.1 Tulu 3 405B

allenai/Llama-3.1-Tulu-3-405B

allenai

[ { "name": "SambaNova", "context": 16000, "max_output": 16000, "input": 5, "output": 10, "latency": 4.53, "throughput": 48.59 } ]

5

10

5

10

5

10

true

DeepSeek: R1 Distill Llama 8B

meta-llama/Llama-3.1-8B

meta-llama

[ { "name": "NovitaAI", "context": 32000, "max_output": 32000, "input": 0.04, "output": 0.04, "latency": 1.11, "throughput": 54.37 } ]

0.04

true

AionLabs: Aion-1.0-Mini

FuseAI/FuseO1-DeepSeekR1-QwQ-SkyT1-32B-Preview

FuseAI

[ { "name": "AionLabs", "context": 33000, "max_output": 33000, "input": 0.7, "output": 1.4, "latency": 0.42, "throughput": 161.4 } ]

0.7

1.4

0.7

1.4

0.7

1.4

true

Qwen: Qwen2.5 VL 72B Instruct

Qwen/Qwen2.5-VL-72B-Instruct

Qwen

[ { "name": "Parasail", "context": 32000, "max_output": 32000, "input": 0.7, "output": 0.7, "latency": 1.78, "throughput": 15.24 } ]

0.7

true

DeepSeek: R1 Distill Qwen 1.5B

Qwen/Qwen2.5-Math-1.5B

Qwen

[ { "name": "Together", "context": 131000, "max_output": 33000, "input": 0.18, "output": 0.18, "latency": 0.26, "throughput": 394.9 } ]

0.18

true

Mistral: Mistral Small 3

mistralai/Mistral-Small-24B-Instruct-2501

mistralai

[ { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.07, "output": 0.14, "latency": 0.7, "throughput": 67.14 }, { "name": "Mistral", "context": 32000, "max_output": 32000, "input": 0.1, "output": 0.3, "latency": 0.33, "throughput": 100.8 }, { "name": "Ubicloud", "context": 33000, "max_output": 33000, "input": 0.3, "output": 0.3, "latency": 1.73, "throughput": 30.95 }, { "name": "Together", "context": 33000, "max_output": 2000, "input": 0.8, "output": 0.8, "latency": 0.47, "throughput": 80.69 }, { "name": "Fireworks", "context": 33000, "max_output": 33000, "input": 0.9, "output": 0.9, "latency": 1.14, "throughput": 33.32 } ]

0.3

0.07

0.14

0.9

true

DeepSeek: R1 Distill Qwen 32B

Qwen/Qwen2.5-32B

Qwen

[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.12, "output": 0.18, "latency": 0.32, "throughput": 43.83 }, { "name": "Cloudflare", "context": 80000, "max_output": 80000, "input": 0.5, "output": 4.88, "latency": 0.42, "throughput": 36 }, { "name": "Groq", "context": 131000, "max_output": 131000, "input": 0.79, "output": 0.79, "latency": 0.13, "throughput": 137.2 }, { "name": "NovitaAI", "context": 64000, "max_output": 64000, "input": 0.3, "output": 0.3, "latency": 8.06, "throughput": 32.05 } ]

0.79

0.12

0.18

0.5

4.88

true

DeepSeek: R1 Distill Qwen 14B

deepseek-ai/DeepSeek-R1-Distill-Qwen-14B

deepseek-ai

[ { "name": "Together", "context": 131000, "max_output": 33000, "input": 1.6, "output": 1.6, "latency": 0.32, "throughput": 163.4 }, { "name": "NovitaAI", "context": 64000, "max_output": 64000, "input": 0.15, "output": 0.15, "latency": 12.9, "throughput": 71.57 } ]

1.6

0.15

1.6

true

DeepSeek: R1 Distill Llama 70B

deepseek-ai/DeepSeek-R1-Distill-Llama-70B

deepseek-ai

[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.23, "output": 0.69, "latency": 10.86, "throughput": 84.41 }, { "name": "Nebius AI Studio", "context": 128000, "max_output": 128000, "input": 0.25, "output": 0.75, "latency": 9.23, "throughput": 64.97 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.4, "output": 0.4, "latency": 19.48, "throughput": 33.32 }, { "name": "SambaNova", "context": 16000, "max_output": 16000, "input": 0.7, "output": 1.4, "latency": 3.8, "throughput": 470.4 }, { "name": "Groq", "context": 131000, "max_output": 131000, "input": 0.75, "output": 0.99, "latency": 1.55, "throughput": 596.1 }, { "name": "Together", "context": 131000, "max_output": 16000, "input": 2, "output": 2, "latency": 5.23, "throughput": 192.7 }, { "name": "NovitaAI", "context": 32000, "max_output": 32000, "input": 0.8, "output": 0.8, "latency": 36.28, "throughput": 33.42 } ]

0.8

0.4

2

true

DeepSeek: R1

deepseek-ai/DeepSeek-R1

deepseek-ai

[ { "name": "Minimax", "context": 64000, "max_output": 64000, "input": 0.55, "output": 2.19, "latency": 2.41, "throughput": 20.27 }, { "name": "Nebius AI Studio", "context": 128000, "max_output": 128000, "input": 0.8, "output": 2.4, "latency": 0.43, "throughput": 12.73 }, { "name": "DeepInfra Turbo", "context": 33000, "max_output": 33000, "input": 2, "output": 6, "latency": 6.53, "throughput": 59.47 }, { "name": "inference.net", "context": 131000, "max_output": 33000, "input": 3, "output": 3, "latency": 0.76, "throughput": 38.14 }, { "name": "Parasail", "context": 128000, "max_output": 128000, "input": 3, "output": 3, "latency": 1.35, "throughput": 55.54 }, { "name": "Together", "context": 164000, "max_output": 8000, "input": 3, "output": 7, "latency": 0.87, "throughput": 58.88 }, { "name": "Friendli", "context": 164000, "max_output": 164000, "input": 3, "output": 7, "latency": 1.21, "throughput": 58.67 }, { "name": "Fireworks", "context": 164000, "max_output": 164000, "input": 3, "output": 8, "latency": 0.81, "throughput": 69.68 }, { "name": "SambaNova", "context": 8000, "max_output": 8000, "input": 5, "output": 7, "latency": 4.69, "throughput": 158.3 }, { "name": "kluster.ai", "context": 128000, "max_output": 128000, "input": 7, "output": 7, "latency": 10.41, "throughput": 34.77 }, { "name": "DeepSeek", "context": 64000, "max_output": 8000, "input": 0.55, "output": 2.19, "latency": 0.69, "throughput": 27.75 }, { "name": "DeepInfra", "context": 66000, "max_output": 8000, "input": 0.75, "output": 2.4, "latency": 2.38, "throughput": 10.05 }, { "name": "NovitaAI", "context": 64000, "max_output": 8000, "input": 4, "output": 4, "latency": 23.74, "throughput": 18.53 }, { "name": "Featherless", "context": 33000, "max_output": 4000, "input": 6.5, "output": 8, "latency": 52.12, "throughput": 19.52 }, { "name": "Avian.io", "context": 164000, "max_output": 164000, "input": 6.9, "output": 6.9, "latency": null, "throughput": null } ]

4

0.55

2.19

6.5

8

true

MiniMax: MiniMax-01

MiniMaxAI/MiniMax-Text-01

MiniMaxAI

[ { "name": "Minimax", "context": 1000000, "max_output": 1000000, "input": 0.2, "output": 1.1, "latency": 1.45, "throughput": 27.07 } ]

0.2

1.1

0.2

1.1

0.2

1.1

true

Microsoft: Phi 4

microsoft/phi-4

microsoft

[ { "name": "DeepInfra", "context": 16000, "max_output": 8000, "input": 0.07, "output": 0.14, "latency": 0.44, "throughput": 37.26 }, { "name": "Nebius AI Studio", "context": 16000, "max_output": 16000, "input": 0.1, "output": 0.3, "latency": 0.14, "throughput": 106 } ]

0.1

0.3

0.07

0.14

0.1

0.3

true

Sao10K: Llama 3.1 70B Hanami x1

Sao10K/L3.1-70B-Hanami-x1

Sao10K

[ { "name": "Infermatic", "context": 16000, "max_output": 16000, "input": 3, "output": 3, "latency": 1.36, "throughput": 30.45 } ]

3

true

DeepSeek: DeepSeek V3

deepseek-ai/DeepSeek-V3

deepseek-ai

[ { "name": "inference.net", "context": 131000, "max_output": 66000, "input": 1.2, "output": 1.2, "latency": 0.89, "throughput": 12.28 }, { "name": "Together", "context": 131000, "max_output": 12000, "input": 1.25, "output": 1.25, "latency": 0.75, "throughput": 34.84 }, { "name": "DeepSeek", "context": 64000, "max_output": 8000, "input": 0.27, "output": 1.1, "latency": 0.48, "throughput": 18.92 }, { "name": "NovitaAI", "context": 64000, "max_output": 16000, "input": 0.4, "output": 1.3, "latency": 1.43, "throughput": 34.49 }, { "name": "DeepInfra", "context": 66000, "max_output": 8000, "input": 0.49, "output": 0.89, "latency": 1.15, "throughput": 7.59 }, { "name": "Nebius AI Studio", "context": 131000, "max_output": 131000, "input": 0.5, "output": 1.5, "latency": 0.3, "throughput": 18.73 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.9, "output": 0.9, "latency": 1.11, "throughput": 49.07 } ]

0.9

0.27

1.1

1.25

true

Sao10K: Llama 3.3 Euryale 70B

Sao10K/L3.3-70B-Euryale-v2.3

Sao10K

[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.7, "output": 0.8, "latency": 0.21, "throughput": 39.36 }, { "name": "Infermatic", "context": 16000, "max_output": 16000, "input": 1.5, "output": 1.5, "latency": 0.73, "throughput": 43.27 } ]

1.5

0.7

0.8

1.5

true

EVA Llama 3.33 70B

EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1

EVA-UNIT-01

[ { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 4, "output": 6, "latency": 3.18, "throughput": 11.5 } ]

4

6

4

6

4

6

true

Meta: Llama 3.3 70B Instruct

meta-llama/Llama-3.3-70B-Instruct

meta-llama

[ { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.12, "output": 0.3, "latency": 0.88, "throughput": 14.05 }, { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.12, "output": 0.3, "latency": 0.39, "throughput": 34.73 }, { "name": "Nebius AI Studio", "context": 131000, "max_output": 131000, "input": 0.13, "output": 0.4, "latency": 0.42, "throughput": 24.18 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.4, "output": 0.4, "latency": 0.56, "throughput": 15.03 }, { "name": "Hyperbolic", "context": 131000, "max_output": 131000, "input": 0.4, "output": 0.4, "latency": 1.37, "throughput": 70.7 }, { "name": "Groq", "context": 33000, "max_output": 33000, "input": 0.59, "output": 0.79, "latency": 0.24, "throughput": 310.7 }, { "name": "Friendli", "context": 131000, "max_output": 131000, "input": 0.6, "output": 0.6, "latency": 0.41, "throughput": 103.5 }, { "name": "SambaNova", "context": 128000, "max_output": 128000, "input": 0.6, "output": 1.2, "latency": 0.9, "throughput": 131.6 }, { "name": "kluster.ai", "context": 131000, "max_output": 131000, "input": 0.7, "output": 0.7, "latency": 0.61, "throughput": 21.99 }, { "name": "Parasail", "context": 131000, "max_output": 131000, "input": 0.7, "output": 0.7, "latency": 0.64, "throughput": 66.03 }, { "name": "Cloudflare", "context": 131000, "max_output": 131000, "input": 0.75, "output": 0.75, "latency": 0.8, "throughput": 28.37 }, { "name": "Lepton", "context": 131000, "max_output": 131000, "input": 0.8, "output": 0.8, "latency": 0.39, "throughput": 47.81 }, { "name": "Together", "context": 131000, "max_output": 2000, "input": 0.88, "output": 0.88, "latency": 0.51, "throughput": 114 }, { "name": "Avian.io", "context": 131000, "max_output": 131000, "input": 0.9, "output": 0.9, "latency": 0.24, "throughput": 122.4 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.9, "output": 0.9, "latency": 0.39, "throughput": 117.1 }, { "name": "NovitaAI", "context": 131000, "max_output": 131000, "input": 0.39, "output": 0.39, "latency": 0.59, "throughput": 44.5 } ]

0.7

0.12

0.3

0.9

true

Qwen: QwQ 32B Preview

Qwen/QwQ-32B-Preview

Qwen

[ { "name": "Hyperbolic", "context": 33000, "max_output": 33000, "input": 0.2, "output": 0.2, "latency": 11.74, "throughput": 35.7 }, { "name": "Fireworks", "context": 33000, "max_output": 33000, "input": 0.9, "output": 0.9, "latency": 5.12, "throughput": 25.3 }, { "name": "Together", "context": 33000, "max_output": 33000, "input": 1.2, "output": 1.2, "latency": 15.19, "throughput": 36.67 }, { "name": "SambaNova", "context": 16000, "max_output": 16000, "input": 1.5, "output": 3, "latency": 0.92, "throughput": 267.7 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.18, "output": 0.6, "latency": null, "throughput": null } ]

0.9

0.2

1.5

3

true

Infermatic: Mistral Nemo Inferor 12B

Infermatic/MN-12B-Inferor-v0.0

Infermatic

[ { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 2.69, "throughput": 14.62 } ]

0.8

1.2

0.8

1.2

0.8

1.2

true

Qwen2.5 Coder 32B Instruct

Qwen/Qwen2.5-Coder-32B-Instruct

Qwen

[ { "name": "Lambda", "context": 33000, "max_output": 3000, "input": 0.07, "output": 0.16, "latency": 0.5, "throughput": 75.8 }, { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.07, "output": 0.16, "latency": 0.26, "throughput": 50.78 }, { "name": "Hyperbolic", "context": 128000, "max_output": 8000, "input": 0.2, "output": 0.2, "latency": 1.14, "throughput": 54.07 }, { "name": "Parasail", "context": 128000, "max_output": 128000, "input": 0.5, "output": 0.5, "latency": 0.56, "throughput": 52.8 }, { "name": "Groq", "context": 131000, "max_output": 131000, "input": 0.79, "output": 0.79, "latency": 0.42, "throughput": 333.3 }, { "name": "Together", "context": 16000, "max_output": 2000, "input": 0.8, "output": 0.8, "latency": 0.5, "throughput": 80.49 }, { "name": "Fireworks", "context": 131000, "max_output": 4000, "input": 0.9, "output": 0.9, "latency": 0.45, "throughput": 55.73 }, { "name": "Mancer", "context": 33000, "max_output": 2000, "input": 1.5, "output": 2.813, "latency": 0.93, "throughput": 19.21 }, { "name": "SambaNova", "context": 16000, "max_output": 16000, "input": 1.5, "output": 3, "latency": 0.78, "throughput": 414.9 }, { "name": "Mancer (private)", "context": 33000, "max_output": 2000, "input": 2, "output": 3.75, "latency": null, "throughput": null }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 2.6, "output": 3.4, "latency": 2.06, "throughput": 13.11 } ]

0.8

0.07

0.16

2.6

3.4

true

SorcererLM 8x22B

rAIfle/SorcererLM-8x22b-bf16

rAIfle

[ { "name": "Infermatic", "context": 16000, "max_output": 16000, "input": 4.5, "output": 4.5, "latency": 2.83, "throughput": 8.29 } ]

4.5

true

EVA Qwen2.5 32B

EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2

EVA-UNIT-01

[ { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 2.6, "output": 3.4, "latency": 1.42, "throughput": 11.58 } ]

2.6

3.4

2.6

3.4

2.6

3.4

true

Unslopnemo 12B

TheDrummer/UnslopNemo-12B-v4.1

TheDrummer

[ { "name": "Infermatic", "context": 32000, "max_output": 32000, "input": 0.5, "output": 0.5, "latency": 0.51, "throughput": 80.43 } ]

0.5

true

NeverSleep: Lumimaid v0.2 70B

NeverSleep/Lumimaid-v0.2-70B

NeverSleep

[ { "name": "Mancer", "context": 16000, "max_output": 2000, "input": 3.375, "output": 4.5, "latency": 3.25, "throughput": 12.05 }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 4, "output": 6, "latency": 2.84, "throughput": 11.9 }, { "name": "Mancer (private)", "context": 16000, "max_output": 2000, "input": 4.5, "output": 6, "latency": 1.77, "throughput": 12.8 } ]

4

6

3.375

4.5

6

true

Magnum v4 72B

anthracite-org/magnum-v4-72b

anthracite-org

[ { "name": "Mancer", "context": 16000, "max_output": 1000, "input": 1.875, "output": 2.25, "latency": 0.87, "throughput": 13.91 }, { "name": "Mancer (private)", "context": 16000, "max_output": 1000, "input": 2.5, "output": 3, "latency": 0.89, "throughput": 13.84 }, { "name": "Infermatic", "context": 33000, "max_output": 33000, "input": 3, "output": 3, "latency": 0.77, "throughput": 22.54 }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 4, "output": 6, "latency": 2.81, "throughput": 11.74 } ]

3

1.875

2.25

4

6

true

Qwen2.5 7B Instruct

Qwen/Qwen2.5-7B-Instruct

Qwen

[ { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.025, "output": 0.05, "latency": 0.19, "throughput": 47.01 }, { "name": "Together", "context": 33000, "max_output": 2000, "input": 0.3, "output": 0.3, "latency": 0.4, "throughput": 123 } ]

0.3

0.025

0.05

0.3

true

NVIDIA: Llama 3.1 Nemotron 70B Instruct

nvidia/Llama-3.1-Nemotron-70B-Instruct-HF

nvidia

[ { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.12, "output": 0.3, "latency": 0.93, "throughput": 33.31 }, { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.12, "output": 0.3, "latency": 0.59, "throughput": 28.13 }, { "name": "Together", "context": 33000, "max_output": 33000, "input": 0.88, "output": 0.88, "latency": 0.54, "throughput": 70.39 }, { "name": "Infermatic", "context": 32000, "max_output": 32000, "input": 1, "output": 1, "latency": 2.63, "throughput": 13.9 } ]

0.88

0.12

0.3

1

true

Magnum v2 72B

anthracite-org/magnum-v2-72b

anthracite-org

[ { "name": "Infermatic", "context": 33000, "max_output": 33000, "input": 3, "output": 3, "latency": 2.98, "throughput": 42.32 }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 4, "output": 6, "latency": 6.41, "throughput": 12.61 } ]

4

6

3

4

6

true

Rocinante 12B

TheDrummer/Rocinante-12B-v1.1

TheDrummer

[ { "name": "Infermatic", "context": 33000, "max_output": 33000, "input": 0.25, "output": 0.5, "latency": 0.65, "throughput": 24.05 }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 2.45, "throughput": 11.45 } ]

0.8

1.2

0.25

0.5

0.8

1.2

true

Meta: Llama 3.2 3B Instruct

meta-llama/Llama-3.2-3B-Instruct

meta-llama

[ { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.015, "output": 0.025, "latency": 0.47, "throughput": 94.12 }, { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.015, "output": 0.025, "latency": 0.08, "throughput": 92.2 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.02, "output": 0.02, "latency": 0.87, "throughput": 85.37 }, { "name": "Lepton", "context": 131000, "max_output": 131000, "input": 0.03, "output": 0.03, "latency": 0.34, "throughput": 201 }, { "name": "Together", "context": 131000, "max_output": 16000, "input": 0.06, "output": 0.06, "latency": 1.08, "throughput": 58.03 }, { "name": "SambaNova", "context": 4000, "max_output": 2000, "input": 0.08, "output": 0.16, "latency": 0.25, "throughput": 1353 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.1, "output": 0.1, "latency": 0.55, "throughput": 156.5 }, { "name": "Hyperbolic", "context": 131000, "max_output": 131000, "input": 0.1, "output": 0.1, "latency": 1.26, "throughput": 187.4 }, { "name": "Cloudflare", "context": 131000, "max_output": 131000, "input": 0.1, "output": 0.1, "latency": 0.89, "throughput": 159.5 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.03, "output": 0.05, "latency": 1.26, "throughput": 79.98 } ]

0.06

0.015

0.025

0.08

0.16

true

Meta: Llama 3.2 1B Instruct

meta-llama/Llama-3.2-1B-Instruct

meta-llama

[ { "name": "Lepton", "context": 131000, "max_output": 131000, "input": 0.01, "output": 0.01, "latency": 0.33, "throughput": 421.3 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.01, "output": 0.01, "latency": 0.53, "throughput": 133.9 }, { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.01, "output": 0.01, "latency": 0.18, "throughput": 154.1 }, { "name": "SambaNova", "context": 4000, "max_output": 2000, "input": 0.04, "output": 0.08, "latency": 0.46, "throughput": 2093 }, { "name": "Cloudflare", "context": 131000, "max_output": 131000, "input": 0.1, "output": 0.1, "latency": 0.87, "throughput": 225.2 }, { "name": "NovitaAI", "context": 131000, "max_output": 131000, "input": 0.02, "output": 0.02, "latency": null, "throughput": null } ]

0.02

0.01

0.1

true

Meta: Llama 3.2 90B Vision Instruct

meta-llama/Llama-3.2-90B-Vision-Instruct

meta-llama

[ { "name": "SambaNova", "context": 4000, "max_output": 2000, "input": 0.8, "output": 1.6, "latency": 0.59, "throughput": 272.8 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.9, "output": 0.9, "latency": 0.86, "throughput": 34.05 }, { "name": "Together", "context": 131000, "max_output": 131000, "input": 1.2, "output": 1.2, "latency": 0.63, "throughput": 35.98 }, { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.35, "output": 0.4, "latency": 0.46, "throughput": 24.82 } ]

1.2

0.35

0.4

0.8

1.6

true

Meta: Llama 3.2 11B Vision Instruct

meta-llama/Llama-3.2-11B-Vision-Instruct

meta-llama

[ { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.055, "output": 0.055, "latency": 0.98, "throughput": 39.34 }, { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.055, "output": 0.055, "latency": 0.14, "throughput": 52.31 }, { "name": "SambaNova", "context": 4000, "max_output": 2000, "input": 0.15, "output": 0.3, "latency": 0.62, "throughput": 869.8 }, { "name": "Together", "context": 131000, "max_output": 131000, "input": 0.18, "output": 0.18, "latency": 0.34, "throughput": 162.1 }, { "name": "Cloudflare", "context": 131000, "max_output": 131000, "input": 0.2, "output": 0.2, "latency": 1.49, "throughput": 35.49 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.2, "output": 0.2, "latency": 5.15, "throughput": 35.08 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.06, "output": 0.06, "latency": null, "throughput": null } ]

0.18

0.055

0.15

0.3

true

Qwen2.5 72B Instruct

Qwen/Qwen2.5-72B-Instruct

Qwen

[ { "name": "Nebius AI Studio", "context": 128000, "max_output": 128000, "input": 0.13, "output": 0.4, "latency": 0.21, "throughput": 29.47 }, { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.13, "output": 0.4, "latency": 0.44, "throughput": 36.79 }, { "name": "Hyperbolic", "context": 33000, "max_output": 33000, "input": 0.4, "output": 0.4, "latency": 1.34, "throughput": 44.11 }, { "name": "Fireworks", "context": 33000, "max_output": 33000, "input": 0.9, "output": 0.9, "latency": 0.39, "throughput": 38.62 }, { "name": "Together", "context": 131000, "max_output": 2000, "input": 1.2, "output": 1.2, "latency": 0.43, "throughput": 96.85 }, { "name": "SambaNova", "context": 16000, "max_output": 16000, "input": 2, "output": 4, "latency": 0.41, "throughput": 359.3 }, { "name": "NovitaAI", "context": 32000, "max_output": 4000, "input": 0.38, "output": 0.4, "latency": 0.83, "throughput": 20.55 } ]

0.4

0.13

0.4

2

4

true

Qwen: Qwen2-VL 72B Instruct

Qwen/Qwen2-VL-72B-Instruct

Qwen

[ { "name": "Hyperbolic", "context": 4000, "max_output": 4000, "input": 0.4, "output": 0.4, "latency": 1.21, "throughput": 37.83 } ]

0.4

true

NeverSleep: Lumimaid v0.2 8B

NeverSleep/Lumimaid-v0.2-8B

NeverSleep

[ { "name": "Mancer", "context": 33000, "max_output": 2000, "input": 0.1875, "output": 1.125, "latency": 0.95, "throughput": 36.34 }, { "name": "Mancer (private)", "context": 33000, "max_output": 2000, "input": 0.25, "output": 1.5, "latency": 0.69, "throughput": 36.51 }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 1.6, "throughput": 26.34 } ]

0.25

1.5

0.1875

1.125

0.8

1.2

true

Mistral: Pixtral 12B

mistralai/Pixtral-12B-2409

mistralai

[ { "name": "Hyperbolic", "context": 4000, "max_output": 4000, "input": 0.1, "output": 0.1, "latency": 1.4, "throughput": 52.28 }, { "name": "Mistral", "context": 4000, "max_output": 4000, "input": 0.15, "output": 0.15, "latency": 1.13, "throughput": 90.11 } ]

0.15

0.1

0.15

true

Sao10K: Llama 3.1 Euryale 70B v2.2

Sao10K/L3.1-70B-Euryale-v2.2

Sao10K

[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.7, "output": 0.8, "latency": 0.32, "throughput": 37.38 }, { "name": "Infermatic", "context": 16000, "max_output": 16000, "input": 1.5, "output": 1.5, "latency": 0.93, "throughput": 19.76 }, { "name": "NovitaAI", "context": 16000, "max_output": 16000, "input": 1.48, "output": 1.48, "latency": 1.51, "throughput": 20.91 } ]

1.48

0.7

0.8

1.5

true

Qwen: Qwen2-VL 7B Instruct

Qwen/Qwen2-VL-7B-Instruct

Qwen

[ { "name": "Hyperbolic", "context": 4000, "max_output": 4000, "input": 0.1, "output": 0.1, "latency": 1, "throughput": 93.31 } ]

0.1

true

Nous: Hermes 3 70B Instruct

NousResearch/Hermes-3-Llama-3.1-70B

NousResearch

[ { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.12, "output": 0.3, "latency": 0.74, "throughput": 32.01 }, { "name": "Hyperbolic", "context": 12000, "max_output": 12000, "input": 0.4, "output": 0.4, "latency": 0.95, "throughput": 31.07 } ]

0.4

0.12

0.3

0.4

true

Nous: Hermes 3 405B Instruct

NousResearch/Hermes-3-Llama-3.1-405B

NousResearch

[ { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.8, "output": 0.8, "latency": 1.1, "throughput": 26.23 }, { "name": "inference.net", "context": 33000, "max_output": 33000, "input": 0.8, "output": 0.8, "latency": 0.92, "throughput": 25.74 }, { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.8, "output": 0.8, "latency": 1.01, "throughput": 11.46 }, { "name": "Nebius AI Studio", "context": 131000, "max_output": 131000, "input": 1, "output": 3, "latency": 0.31, "throughput": 26.61 } ]

0.8

1

3

true

Sao10K: Llama 3 8B Lunaris

Sao10K/L3-8B-Lunaris-v1

Sao10K

[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.03, "output": 0.06, "latency": 0.38, "throughput": 73.74 }, { "name": "NovitaAI", "context": 8000, "max_output": 8000, "input": 0.05, "output": 0.05, "latency": 0.88, "throughput": 59.26 } ]

0.05

0.03

0.06

0.05

true

Aetherwiing: Starcannon 12B

intervitens/mini-magnum-12b-v1.1

intervitens

[ { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 3.92, "throughput": 15.83 } ]

0.8

1.2

0.8

1.2

0.8

1.2

true

Meta: Llama 3.1 405B (base)

meta-llama/llama-3.1-405B

meta-llama

[ { "name": "Hyperbolic (quantized)", "context": 33000, "max_output": 33000, "input": 2, "output": 2, "latency": 0.74, "throughput": 23.31 }, { "name": "Hyperbolic", "context": 33000, "max_output": 33000, "input": 4, "output": 4, "latency": 1.19, "throughput": 16.15 } ]

4

2

4

true

Mistral Nemo 12B Celeste

nothingiisreal/MN-12B-Celeste-V1.9

nothingiisreal

[ { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 2.54, "throughput": 14.35 } ]

0.8

1.2

0.8

1.2

0.8

1.2

true

Meta: Llama 3.1 405B Instruct

meta-llama/Meta-Llama-3.1-405B-Instruct

meta-llama

[ { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.8, "output": 0.8, "latency": 0.15, "throughput": 17.38 }, { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.8, "output": 0.8, "latency": 0.64, "throughput": 34.89 }, { "name": "Nebius AI Studio", "context": 128000, "max_output": 128000, "input": 1, "output": 3, "latency": 0.15, "throughput": 27.72 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 3, "output": 3, "latency": 1.12, "throughput": 44.58 }, { "name": "Together", "context": 131000, "max_output": 131000, "input": 3.5, "output": 3.5, "latency": 1.08, "throughput": 49.02 }, { "name": "kluster.ai", "context": 131000, "max_output": 131000, "input": 3.5, "output": 3.5, "latency": 1.41, "throughput": 17.25 }, { "name": "Hyperbolic", "context": 33000, "max_output": 33000, "input": 4, "output": 4, "latency": 1.02, "throughput": 8.78 }, { "name": "SambaNova", "context": 8000, "max_output": 4000, "input": 5, "output": 10, "latency": 1.53, "throughput": 81.81 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 2.75, "output": 2.75, "latency": null, "throughput": null } ]

3

0.8

5

10

true

Meta: Llama 3.1 8B Instruct

meta-llama/Meta-Llama-3.1-8B-Instruct

meta-llama

[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.02, "output": 0.05, "latency": 0.08, "throughput": 74.22 }, { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.025, "output": 0.04, "latency": 0.43, "throughput": 220.6 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.03, "output": 0.03, "latency": 0.63, "throughput": 59.25 }, { "name": "Groq", "context": 131000, "max_output": 8000, "input": 0.05, "output": 0.08, "latency": 0.3, "throughput": 552.2 }, { "name": "Lepton", "context": 131000, "max_output": 131000, "input": 0.07, "output": 0.07, "latency": 0.21, "throughput": 100.1 }, { "name": "Friendli", "context": 131000, "max_output": 131000, "input": 0.1, "output": 0.1, "latency": 0.42, "throughput": 245.1 }, { "name": "Hyperbolic", "context": 33000, "max_output": 33000, "input": 0.1, "output": 0.1, "latency": 1.19, "throughput": 101 }, { "name": "SambaNova", "context": 8000, "max_output": 4000, "input": 0.1, "output": 0.2, "latency": 0.28, "throughput": 803.3 }, { "name": "Cloudflare", "context": 131000, "max_output": 131000, "input": 0.15, "output": 0.15, "latency": 0.31, "throughput": 126.3 }, { "name": "kluster.ai", "context": 131000, "max_output": 131000, "input": 0.18, "output": 0.18, "latency": 0.32, "throughput": 13.41 }, { "name": "Together", "context": 131000, "max_output": 131000, "input": 0.18, "output": 0.18, "latency": 0.25, "throughput": 313.2 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.2, "output": 0.2, "latency": 0.36, "throughput": 223.2 }, { "name": "Avian.io", "context": 131000, "max_output": 131000, "input": 0.2, "output": 0.2, "latency": 0.29, "throughput": 308 }, { "name": "NovitaAI", "context": 16000, "max_output": 8000, "input": 0.05, "output": 0.05, "latency": 1.84, "throughput": 66.66 } ]

0.1

0.03

0.2

true

Meta: Llama 3.1 70B Instruct

meta-llama/Meta-Llama-3.1-70B-Instruct

meta-llama

[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.12, "output": 0.3, "latency": 0.54, "throughput": 33.04 }, { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.12, "output": 0.3, "latency": 0.58, "throughput": 37.2 }, { "name": "Nebius AI Studio", "context": 128000, "max_output": 128000, "input": 0.13, "output": 0.4, "latency": 0.15, "throughput": 29.24 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.4, "output": 0.4, "latency": 1.38, "throughput": 16.81 }, { "name": "Hyperbolic", "context": 33000, "max_output": 33000, "input": 0.4, "output": 0.4, "latency": 0.88, "throughput": 106.8 }, { "name": "Friendli", "context": 131000, "max_output": 131000, "input": 0.6, "output": 0.6, "latency": 0.2, "throughput": 113.5 }, { "name": "SambaNova", "context": 8000, "max_output": 4000, "input": 0.6, "output": 1.2, "latency": 0.53, "throughput": 269.4 }, { "name": "Cloudflare", "context": 131000, "max_output": 131000, "input": 0.75, "output": 0.75, "latency": 0.78, "throughput": 27.26 }, { "name": "Lepton", "context": 131000, "max_output": 131000, "input": 0.8, "output": 0.8, "latency": 0.3, "throughput": 45.24 }, { "name": "Together", "context": 131000, "max_output": 131000, "input": 0.88, "output": 0.88, "latency": 0.33, "throughput": 115.5 }, { "name": "Fireworks", "context": 131000, "max_output": 131000, "input": 0.9, "output": 0.9, "latency": 0.44, "throughput": 94.56 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.34, "output": 0.39, "latency": 1.72, "throughput": 52.16 } ]

0.6

0.12

0.3

0.9

true

Mistral: Mistral Nemo

mistralai/Mistral-Nemo-Instruct-2407

mistralai

[ { "name": "DeepInfra", "context": 131000, "max_output": 8000, "input": 0.035, "output": 0.08, "latency": 0.25, "throughput": 52.49 }, { "name": "Nebius AI Studio", "context": 128000, "max_output": 128000, "input": 0.04, "output": 0.12, "latency": 0.64, "throughput": 22.83 }, { "name": "inference.net", "context": 16000, "max_output": 16000, "input": 0.1, "output": 0.1, "latency": 0.69, "throughput": 63.6 }, { "name": "Parasail", "context": 128000, "max_output": 128000, "input": 0.11, "output": 0.11, "latency": 0.75, "throughput": 134.1 }, { "name": "Mistral", "context": 128000, "max_output": 128000, "input": 0.15, "output": 0.15, "latency": 0.29, "throughput": 128.1 }, { "name": "Lepton", "context": 128000, "max_output": 128000, "input": 0.18, "output": 0.18, "latency": 0.28, "throughput": 50.18 }, { "name": "Azure", "context": 128000, "max_output": 128000, "input": 0.3, "output": 0.3, "latency": 0.64, "throughput": 90.2 }, { "name": "NovitaAI", "context": 131000, "max_output": 131000, "input": 0.17, "output": 0.17, "latency": 1.61, "throughput": 39.52 } ]

0.15

0.035

0.08

0.3

true

Mistral: Codestral Mamba

mistralai/mamba-codestral-7B-v0.1

mistralai

[ { "name": "Mistral", "context": 256000, "max_output": 256000, "input": 0.25, "output": 0.25, "latency": 0.34, "throughput": 97.73 } ]

0.25

true

Qwen 2 7B Instruct

Qwen/Qwen2-7B-Instruct

Qwen

[ { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.054, "output": 0.054, "latency": 0.65, "throughput": 82.28 } ]

0.054

true

Google: Gemma 2 27B

google/gemma-2-27b-it

google

[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.27, "output": 0.27, "latency": 0.49, "throughput": 38.62 }, { "name": "Together", "context": 8000, "max_output": 2000, "input": 0.8, "output": 0.8, "latency": 0.41, "throughput": 69.52 } ]

0.8

0.27

0.8

true

Magnum 72B

alpindale/magnum-72b-v1

alpindale

[ { "name": "Mancer", "context": 16000, "max_output": 1000, "input": 1.875, "output": 2.25, "latency": 1.21, "throughput": 22.84 }, { "name": "Mancer (private)", "context": 16000, "max_output": 1000, "input": 2.5, "output": 3, "latency": 0.92, "throughput": 22.6 }, { "name": "Featherless", "context": 16000, "max_output": 4000, "input": 4, "output": 6, "latency": 8.11, "throughput": 11.31 } ]

2.5

3

1.875

2.25

4

6

true

Google: Gemma 2 9B

google/gemma-2-9b-it

google

[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.03, "output": 0.06, "latency": 0.09, "throughput": 35.94 }, { "name": "Lepton", "context": 8000, "max_output": 8000, "input": 0.07, "output": 0.07, "latency": 0.27, "throughput": 103.4 }, { "name": "Groq", "context": 8000, "max_output": 8000, "input": 0.2, "output": 0.2, "latency": 0.22, "throughput": 560.1 }, { "name": "Together", "context": 8000, "max_output": 8000, "input": 0.3, "output": 0.3, "latency": 0.22, "throughput": 109 }, { "name": "NovitaAI", "context": 8000, "max_output": 8000, "input": 0.08, "output": 0.08, "latency": 4.21, "throughput": 29.98 } ]

0.08

0.03

0.06

0.3

true

Sao10k: Llama 3 Euryale 70B v2.1

Sao10K/L3-70B-Euryale-v2.1

Sao10K

[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.7, "output": 0.8, "latency": 0.2, "throughput": 40.32 }, { "name": "NovitaAI", "context": 16000, "max_output": 16000, "input": 1.48, "output": 1.48, "latency": 3.58, "throughput": 19.49 } ]

1.48

0.7

0.8

1.48

true

Dolphin 2.9.2 Mixtral 8x22B 🐬

cognitivecomputations/dolphin-2.9.2-mixtral-8x22b

cognitivecomputations

[ { "name": "NovitaAI", "context": 16000, "max_output": 16000, "input": 0.9, "output": 0.9, "latency": 6.55, "throughput": 9.32 } ]

0.9

true

Qwen 2 72B Instruct

Qwen/Qwen2-72B-Instruct

Qwen

[ { "name": "Together", "context": 33000, "max_output": 4000, "input": 0.9, "output": 0.9, "latency": 0.45, "throughput": 67.28 } ]

0.9

true

Mistral: Mistral 7B Instruct

mistralai/Mistral-7B-Instruct-v0.3

mistralai

[ { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.03, "output": 0.055, "latency": 0.29, "throughput": 82.05 }, { "name": "Parasail", "context": 33000, "max_output": 33000, "input": 0.11, "output": 0.11, "latency": 0.63, "throughput": 128.1 }, { "name": "Together", "context": 33000, "max_output": 4000, "input": 0.2, "output": 0.2, "latency": 0.37, "throughput": 163.2 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.059, "output": 0.059, "latency": 1.1, "throughput": 121.5 }, { "name": "Lepton", "context": 33000, "max_output": 33000, "input": 0.07, "output": 0.07, "latency": 0.54, "throughput": 105.4 } ]

0.07

0.03

0.055

0.2

true

Mistral: Mistral 7B Instruct v0.3

mistralai/Mistral-7B-Instruct-v0.3

mistralai

[ { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.03, "output": 0.055, "latency": 0.2, "throughput": 91.8 }, { "name": "Lepton", "context": 33000, "max_output": 33000, "input": 0.07, "output": 0.07, "latency": 0.28, "throughput": 107.9 }, { "name": "Together", "context": 33000, "max_output": 4000, "input": 0.2, "output": 0.2, "latency": 0.46, "throughput": 136.8 }, { "name": "NovitaAI", "context": 33000, "max_output": 33000, "input": 0.059, "output": 0.059, "latency": null, "throughput": null } ]

0.07

0.03

0.055

0.2

true

NousResearch: Hermes 2 Pro - Llama-3 8B

NousResearch/Hermes-2-Pro-Llama-3-8B

NousResearch

[ { "name": "Lambda", "context": 131000, "max_output": 131000, "input": 0.025, "output": 0.04, "latency": 0.5, "throughput": 271.2 }, { "name": "NovitaAI", "context": 8000, "max_output": 8000, "input": 0.14, "output": 0.14, "latency": 1.23, "throughput": 126.7 } ]

0.14

0.025

0.04

0.14

true

Microsoft: Phi-3 Mini 128K Instruct

microsoft/Phi-3-mini-128k-instruct

microsoft

[ { "name": "Azure", "context": 128000, "max_output": 128000, "input": 0.1, "output": 0.1, "latency": 0.48, "throughput": 91.92 } ]

0.1

true

Microsoft: Phi-3 Medium 128K Instruct

microsoft/Phi-3-medium-128k-instruct

microsoft

[ { "name": "Azure", "context": 128000, "max_output": 128000, "input": 1, "output": 1, "latency": 0.56, "throughput": 58.03 } ]

1

true

NeverSleep: Llama 3 Lumimaid 70B

NeverSleep/Llama-3-Lumimaid-70B-v0.1

NeverSleep

[ { "name": "Mancer", "context": 8000, "max_output": 2000, "input": 3.375, "output": 4.5, "latency": 0.91, "throughput": 13.09 }, { "name": "Featherless", "context": 8000, "max_output": 4000, "input": 4, "output": 6, "latency": 2.25, "throughput": 9.73 }, { "name": "Mancer (private)", "context": 8000, "max_output": 2000, "input": 4.5, "output": 6, "latency": 0.82, "throughput": 13.63 } ]

4

6

3.375

4.5

6

true

Meta: LlamaGuard 2 8B

meta-llama/LlamaGuard-7b

meta-llama

[ { "name": "Together", "context": 8000, "max_output": 8000, "input": 0.2, "output": 0.2, "latency": 0.98, "throughput": 76.25 } ]

0.2

true

NeverSleep: Llama 3 Lumimaid 8B (extended)

NeverSleep/Llama-3-Lumimaid-8B-v0.1

NeverSleep

[ { "name": "Mancer", "context": 25000, "max_output": 2000, "input": 0.1875, "output": 1.125, "latency": 0.77, "throughput": 50.08 }, { "name": "Mancer (private)", "context": 25000, "max_output": 2000, "input": 0.25, "output": 1.5, "latency": 0.54, "throughput": 53.36 } ]

0.25

1.5

0.1875

1.125

0.25

1.5

true

NeverSleep: Llama 3 Lumimaid 8B

NeverSleep/Llama-3-Lumimaid-8B-v0.1

NeverSleep

[ { "name": "Mancer", "context": 25000, "max_output": 2000, "input": 0.1875, "output": 1.125, "latency": 0.76, "throughput": 50.01 }, { "name": "Mancer (private)", "context": 25000, "max_output": 2000, "input": 0.25, "output": 1.5, "latency": 0.54, "throughput": 53.36 }, { "name": "Featherless", "context": 8000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 1.3, "throughput": 25.6 } ]

0.25

1.5

0.1875

1.125

0.8

1.2

true

Fimbulvetr 11B v2

Sao10K/Fimbulvetr-11B-v2

Sao10K

[ { "name": "Featherless", "context": 4000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 1.74, "throughput": 24.72 } ]

0.8

1.2

0.8

1.2

0.8

1.2

true

Meta: Llama 3 8B Instruct

meta-llama/Meta-Llama-3-8B-Instruct

meta-llama

[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.03, "output": 0.06, "latency": 0.17, "throughput": 115.6 }, { "name": "Groq", "context": 8000, "max_output": 8000, "input": 0.05, "output": 0.08, "latency": 0.29, "throughput": 478.4 }, { "name": "Cloudflare", "context": 8000, "max_output": 8000, "input": 0.15, "output": 0.15, "latency": 0.72, "throughput": 15.63 }, { "name": "Mancer", "context": 16000, "max_output": 2000, "input": 0.1875, "output": 1.125, "latency": 0.56, "throughput": 51.09 }, { "name": "Fireworks", "context": 8000, "max_output": 8000, "input": 0.2, "output": 0.2, "latency": 0.48, "throughput": 146.5 }, { "name": "Mancer (private)", "context": 16000, "max_output": 2000, "input": 0.25, "output": 1.5, "latency": 0.6, "throughput": 51.02 }, { "name": "NovitaAI", "context": 8000, "max_output": 8000, "input": 0.04, "output": 0.04, "latency": null, "throughput": null } ]

0.15

0.04

0.25

1.5

true

Meta: Llama 3 70B Instruct

meta-llama/Meta-Llama-3-70B-Instruct

meta-llama

[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.23, "output": 0.4, "latency": 0.41, "throughput": 24.94 }, { "name": "Groq", "context": 8000, "max_output": 8000, "input": 0.59, "output": 0.79, "latency": 0.22, "throughput": 265 }, { "name": "Together", "context": 8000, "max_output": 8000, "input": 0.88, "output": 0.88, "latency": 0.42, "throughput": 20.15 }, { "name": "Fireworks", "context": 8000, "max_output": 8000, "input": 0.9, "output": 0.9, "latency": 0.59, "throughput": 162.2 }, { "name": "NovitaAI", "context": 8000, "max_output": 8000, "input": 0.51, "output": 0.74, "latency": 4.84, "throughput": 20.38 } ]

0.59

0.79

0.23

0.4

0.9

true

Mistral: Mixtral 8x22B Instruct

mistralai/Mixtral-8x22B-Instruct-v0.1

mistralai

[ { "name": "Fireworks", "context": 66000, "max_output": 66000, "input": 0.9, "output": 0.9, "latency": 0.67, "throughput": 77.48 }, { "name": "Together", "context": 66000, "max_output": 2000, "input": 1.2, "output": 1.2, "latency": 0.78, "throughput": 104.2 }, { "name": "Mistral", "context": 66000, "max_output": 66000, "input": 2, "output": 6, "latency": 0.26, "throughput": 97.77 } ]

1.2

0.9

2

6

true

WizardLM-2 8x22B

microsoft/WizardLM-2-8x22B

microsoft

[ { "name": "DeepInfra", "context": 66000, "max_output": 8000, "input": 0.5, "output": 0.5, "latency": 0.1, "throughput": 29.57 }, { "name": "Lepton", "context": 66000, "max_output": 66000, "input": 1, "output": 1, "latency": 0.39, "throughput": 33.75 }, { "name": "NovitaAI", "context": 66000, "max_output": 66000, "input": 0.62, "output": 0.62, "latency": 1.06, "throughput": 17.49 }, { "name": "Together", "context": 66000, "max_output": 66000, "input": 1.2, "output": 1.2, "latency": 0.56, "throughput": 64.35 } ]

1

0.5

1.2

true

WizardLM-2 7B

microsoft/WizardLM-2-7B

microsoft

[ { "name": "Lepton", "context": 32000, "max_output": 32000, "input": 0.07, "output": 0.07, "latency": 0.36, "throughput": 102.4 } ]

0.07

true

Databricks: DBRX 132B Instruct

databricks/dbrx-instruct

databricks

[ { "name": "Together", "context": 33000, "max_output": 2000, "input": 1.2, "output": 1.2, "latency": 0.42, "throughput": 83.57 } ]

1.2

true

Midnight Rose 70B

sophosympatheia/Wizard-Tulu-Dolphin-70B-v1.0

sophosympatheia

[ { "name": "NovitaAI", "context": 4000, "max_output": 4000, "input": 0.8, "output": 0.8, "latency": 1.84, "throughput": 14.85 } ]

0.8

true

Google: Gemma 7B

google/gemma-1.1-7b-it

google

[ { "name": "Cloudflare", "context": 8000, "max_output": 8000, "input": 0.15, "output": 0.15, "latency": 1.33, "throughput": 13.07 } ]

0.15

true

Nous: Hermes 2 Mixtral 8x7B DPO

NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO

NousResearch

[ { "name": "Together", "context": 33000, "max_output": 2000, "input": 0.6, "output": 0.6, "latency": 0.45, "throughput": 105.8 } ]

0.6

true

Dolphin 2.6 Mixtral 8x7B 🐬

cognitivecomputations/dolphin-2.6-mixtral-8x7b

cognitivecomputations

[ { "name": "Lepton", "context": 33000, "max_output": 33000, "input": 0.5, "output": 0.5, "latency": 0.11, "throughput": 84.13 } ]

0.5

true

Mistral: Mixtral 8x7B (base)

mistralai/Mixtral-8x7B-v0.1

mistralai

[ { "name": "Together", "context": 33000, "max_output": 2000, "input": 0.6, "output": 0.6, "latency": 0.38, "throughput": 128 } ]

0.6

true

Mistral: Mixtral 8x7B Instruct

mistralai/Mixtral-8x7B-Instruct-v0.1

mistralai

[ { "name": "DeepInfra", "context": 33000, "max_output": 8000, "input": 0.24, "output": 0.24, "latency": 0.31, "throughput": 101.4 }, { "name": "Groq", "context": 33000, "max_output": 33000, "input": 0.24, "output": 0.24, "latency": 0.34, "throughput": 676.8 }, { "name": "Fireworks", "context": 33000, "max_output": 33000, "input": 0.5, "output": 0.5, "latency": 0.62, "throughput": 155.2 }, { "name": "Lepton", "context": 33000, "max_output": 33000, "input": 0.5, "output": 0.5, "latency": 0.31, "throughput": 73.77 }, { "name": "Together", "context": 33000, "max_output": 2000, "input": 0.6, "output": 0.6, "latency": 0.51, "throughput": 92.37 } ]

0.5

0.24

0.6

true

OpenChat 3.5 7B

openchat/openchat-3.5-0106

openchat

[ { "name": "DeepInfra", "context": 8000, "max_output": 8000, "input": 0.055, "output": 0.055, "latency": 0.23, "throughput": 98.01 }, { "name": "Lepton", "context": 8000, "max_output": 8000, "input": 0.07, "output": 0.07, "latency": 0.41, "throughput": 106.5 }, { "name": "NovitaAI", "context": 4000, "max_output": 4000, "input": 0.06, "output": 0.06, "latency": 0.8, "throughput": 58.78 } ]

0.06

0.055

0.07

true

Noromaid 20B

NeverSleep/Noromaid-20b-v0.1.1

NeverSleep

[ { "name": "Mancer", "context": 8000, "max_output": 2000, "input": 1.5, "output": 2.25, "latency": 0.55, "throughput": 26.36 }, { "name": "Mancer (private)", "context": 8000, "max_output": 2000, "input": 2, "output": 3, "latency": 1.05, "throughput": 25.55 } ]

2

3

1.5

2.25

2

3

true

OpenHermes 2.5 Mistral 7B

teknium/OpenHermes-2.5-Mistral-7B

teknium

[ { "name": "NovitaAI", "context": 4000, "max_output": 4000, "input": 0.17, "output": 0.17, "latency": 0.66, "throughput": 148.9 } ]

0.17

true

Toppy M 7B

Undi95/Toppy-M-7B

Undi95

[ { "name": "Lepton", "context": 4000, "max_output": 4000, "input": 0.07, "output": 0.07, "latency": 0.41, "throughput": 107.1 } ]

0.07

true

Goliath 120B

alpindale/goliath-120b

alpindale

[ { "name": "Mancer", "context": 6000, "max_output": 512, "input": 9.375, "output": 9.375, "latency": 1.44, "throughput": 17.83 }, { "name": "Mancer (private)", "context": 6000, "max_output": 512, "input": 12.5, "output": 12.5, "latency": 1.29, "throughput": 15.68 } ]

12.5

9.375

12.5

true

Airoboros 70B

jondurbin/airoboros-l2-70b-2.2.1

jondurbin

[ { "name": "NovitaAI", "context": 4000, "max_output": 4000, "input": 0.5, "output": 0.5, "latency": 2.14, "throughput": 56.12 } ]

0.5

true

Xwin 70B

Xwin-LM/Xwin-LM-70B-V0.1

Xwin-LM

[ { "name": "Mancer", "context": 8000, "max_output": 512, "input": 3.75, "output": 3.75, "latency": 2.52, "throughput": 18.03 }, { "name": "Mancer (private)", "context": 8000, "max_output": 512, "input": 5, "output": 5, "latency": 2.09, "throughput": 18.19 } ]

5

3.75

5

true

Mistral: Mistral 7B Instruct v0.1

mistralai/Mistral-7B-Instruct-v0.1

mistralai

[ { "name": "Together", "context": 33000, "max_output": 2000, "input": 0.2, "output": 0.2, "latency": 0.31, "throughput": 178.2 } ]

0.2

true

Pygmalion: Mythalion 13B

PygmalionAI/mythalion-13b

PygmalionAI

[ { "name": "Featherless", "context": 4000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": 1.86, "throughput": 12.32 }, { "name": "Mancer", "context": 8000, "max_output": 512, "input": 1.125, "output": 1.125, "latency": 0.92, "throughput": 22.86 }, { "name": "Mancer (private)", "context": 8000, "max_output": 512, "input": 1.5, "output": 1.5, "latency": 0.74, "throughput": 22.35 } ]

1.125

0.8

1.2

1.5

true

Nous: Hermes 13B

NousResearch/Nous-Hermes-Llama2-13b

NousResearch

[ { "name": "Lepton", "context": 4000, "max_output": 4000, "input": 0.18, "output": 0.18, "latency": 0.26, "throughput": 81.6 }, { "name": "NovitaAI", "context": 4000, "max_output": 4000, "input": 0.17, "output": 0.17, "latency": 0.83, "throughput": 82.68 } ]

0.18

0.17

0.18

true

ReMM SLERP 13B

Undi95/ReMM-SLERP-L2-13B

Undi95

[ { "name": "Featherless", "context": 4000, "max_output": 4000, "input": 0.8, "output": 1.2, "latency": null, "throughput": null }, { "name": "Mancer", "context": 6000, "max_output": 512, "input": 1.125, "output": 1.125, "latency": 0.6, "throughput": 43.76 }, { "name": "Mancer (private)", "context": 6000, "max_output": 512, "input": 1.5, "output": 1.5, "latency": 0.57, "throughput": 43.3 } ]

1.125

0.8

1.2

1.5

true

MythoMax 13B

Gryphe/MythoMax-L2-13b

Gryphe

[ { "name": "DeepInfra", "context": 4000, "max_output": 4000, "input": 0.065, "output": 0.065, "latency": 0.2, "throughput": 108.8 }, { "name": "Parasail", "context": 4000, "max_output": 4000, "input": 0.11, "output": 0.11, "latency": 0.56, "throughput": 91.29 }, { "name": "Lepton", "context": 4000, "max_output": 4000, "input": 0.18, "output": 0.18, "latency": 0.21, "throughput": 104.1 }, { "name": "Fireworks", "context": 4000, "max_output": 4000, "input": 0.2, "output": 0.2, "latency": 2.02, "throughput": 7.59 }, { "name": "Together", "context": 4000, "max_output": 4000, "input": 0.3, "output": 0.3, "latency": 0.45, "throughput": 123.7 }, { "name": "Mancer", "context": 8000, "max_output": 512, "input": 1.125, "output": 1.125, "latency": 0.59, "throughput": 41.95 }, { "name": "Mancer (private)", "context": 8000, "max_output": 512, "input": 1.5, "output": 1.5, "latency": 0.92, "throughput": 41.69 }, { "name": "NovitaAI", "context": 4000, "max_output": 4000, "input": 0.09, "output": 0.09, "latency": 1.9, "throughput": 79.17 } ]

0.2

0.065

1.5

true

Meta: Llama 2 13B Chat

meta-llama/Llama-2-13b-chat-hf

meta-llama

[ { "name": "Together", "context": 4000, "max_output": 2000, "input": 0.22, "output": 0.22, "latency": 0.76, "throughput": 63.12 }, { "name": "Lepton", "context": 4000, "max_output": 4000, "input": 0.3, "output": 0.3, "latency": 0.44, "throughput": 98.15 } ]

0.3

0.22

0.3

true

Meta: Llama 2 70B Chat

meta-llama/Llama-2-70b-chat-hf

meta-llama

[ { "name": "Together", "context": 4000, "max_output": 4000, "input": 0.9, "output": 0.9, "latency": 0.73, "throughput": 45.19 } ]

0.9

true

Perplexity: Sonar Reasoning Pro

null

perplexity-ai

[ { "name": "Perplexity", "context": 128000, "max_output": 128000, "input": 2, "output": 8, "latency": 11.05, "throughput": 91.82 } ]

2

8

2

8

2

8

false

Perplexity: Sonar Pro

null

perplexity-ai

[ { "name": "Perplexity", "context": 200000, "max_output": 8000, "input": 3, "output": 15, "latency": 2.49, "throughput": 51.06 } ]

3

15

3

15

3

15

false

Perplexity: Sonar Deep Research

null

perplexity-ai

[ { "name": "Perplexity", "context": 200000, "max_output": 200000, "input": 2, "output": 8, "latency": 24.87, "throughput": 51.64 } ]

2

8

2

8

2

8

false

OpenAI: GPT-4.5 (Preview)

null

OpenAI

[ { "name": "OpenAI", "context": 128000, "max_output": 16000, "input": 75, "output": 150, "latency": 1.64, "throughput": 14.34 } ]

75

150

75

150

75

150

false

Datasets:

cfahlgren1
/

open-inference-pricing

Open Inference Pricing 💰