[CPU] Support FP8 KV cache (#14741)

Signed-off-by: jiang1.li <jiang1.li@intel.com>
This commit is contained in:
Li, Jiang
2025-03-15 13:07:36 +08:00
committed by GitHub
parent 877e352262
commit a2ae496589
8 changed files with 122 additions and 36 deletions

View File

@@ -189,7 +189,7 @@ vLLM CPU backend supports the following vLLM features:
- Model Quantization (`INT8 W8A8, AWQ, GPTQ`)
- Chunked-prefill
- Prefix-caching
- FP8-E5M2 KV-Caching (TODO)
- FP8-E5M2 KV cache
## Related runtime environment variables