This website requires JavaScript.
Explore
Help
Sign In
iHore
/
vLLM
Watch
8
Star
0
Fork
0
Code
Issues
Pull Requests
Actions
2
Packages
Projects
Releases
Wiki
Activity
Files
382970ee6ca490aeaaaf4e32c53695b581ff61ba
vLLM
/
tests
/
models
/
language
/
generation_ppl_test
T
History
music-dino
a9b39d6dcb
[Bugfix] Enable chunked prefill for qwen3.5-0.8B ppl test (
#51153
)
...
Signed-off-by: Dino Music <
Dino.Music@amd.com
>
2026-08-05 13:21:43 +00:00
..
__init__.py
[CI] Add PPL test for generation models (
#24485
)
2025-09-10 06:16:39 -07:00
ppl_utils.py
[Core] Support fp32 lm_head for generation models via head_dtype (RFC
#48305
§3.6) (
#48390
)
2026-07-13 16:43:34 +08:00
test_gemma.py
[CI] And PPL test for Qwen3.5. (
#35853
)
2026-03-03 13:15:51 +00:00
test_gpt.py
[Core] Support fp32 lm_head for generation models via head_dtype (RFC
#48305
§3.6) (
#48390
)
2026-07-13 16:43:34 +08:00
test_qwen.py
[Bugfix] Enable chunked prefill for qwen3.5-0.8B ppl test (
#51153
)
2026-08-05 13:21:43 +00:00