Has anyone evaluated the performance of the AWQ version of the model on benchmarks?
#8
by
liuqianchao
- opened
In my own task benchmark tests, under the condition of long prompt inputs, the performance of the AWQ model is significantly inferior to the 671B full parameter version. Has anyone else compared the performance?
What is your benchmark result? @liuqianchao
+1
Any update？
v2ray
changed discussion status to
closed