biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Model] Revert PR #26715: Restore custom PaliGemma and Gemma3-MM impl… (#27309)

Browse Source 
Signed-off-by: Luciano Martins <lucianommartins@users.noreply.github.com>
Co-authored-by: Luciano Martins <lucianommartins@users.noreply.github.com>
This commit is contained in:
Luciano Martins
2025-10-22 14:05:34 -03:00
committed by GitHub
parent 084a9dae80
commit e05a6754a8
12 changed files with 1219 additions and 54 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
vllm/platforms/rocm.py
View File

@@ -59,6 +59,9 @@ _ROCM_PARTIALLY_SUPPORTED_MODELS: dict[str, str] = {
"Qwen2ForCausalLM": _ROCM_SWA_REASON,
"MistralForCausalLM": _ROCM_SWA_REASON,
"MixtralForCausalLM": _ROCM_SWA_REASON,
"PaliGemmaForConditionalGeneration": (
"ROCm flash attention does not yet fully support 32-bit precision on PaliGemma"
),
"Phi3VForCausalLM": (
"ROCm Triton flash attention may run into compilation errors due to "
"excessive use of shared memory. If this happens, disable Triton FA "