[CI] Disable CUDA graphs for GLM PCP evals (#54745)
Signed-off-by: Kevin Luu <51931015+khluu@users.noreply.github.com>
This commit is contained in:
@@ -4,7 +4,7 @@ num_questions: 1319
|
|||||||
num_fewshot: 5
|
num_fewshot: 5
|
||||||
max_concurrency: 100
|
max_concurrency: 100
|
||||||
server_args: >-
|
server_args: >-
|
||||||
--compilation-config '{"cudagraph_mode":"PIECEWISE"}'
|
--compilation-config '{"cudagraph_mode":"NONE"}'
|
||||||
--max-model-len 4096
|
--max-model-len 4096
|
||||||
--max-num-batched-tokens 32768
|
--max-num-batched-tokens 32768
|
||||||
--safetensors-load-strategy prefetch
|
--safetensors-load-strategy prefetch
|
||||||
|
|||||||
@@ -4,7 +4,7 @@ num_questions: 1319
|
|||||||
num_fewshot: 5
|
num_fewshot: 5
|
||||||
max_concurrency: 100
|
max_concurrency: 100
|
||||||
server_args: >-
|
server_args: >-
|
||||||
--compilation-config '{"cudagraph_mode":"PIECEWISE"}'
|
--compilation-config '{"cudagraph_mode":"NONE"}'
|
||||||
--max-model-len 4096
|
--max-model-len 4096
|
||||||
--max-num-batched-tokens 32768
|
--max-num-batched-tokens 32768
|
||||||
--safetensors-load-strategy prefetch
|
--safetensors-load-strategy prefetch
|
||||||
|
|||||||
Reference in New Issue
Block a user