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

[Model][6/N] Improve all pooling task | Support chunked prefill with ALL pooling (#27145)

Browse Source 
Signed-off-by: wang.yuqi <noooop@126.com>
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
This commit is contained in:
wang.yuqi
2025-12-04 21:44:15 +08:00
committed by GitHub
parent 1b7c7f5159
commit 74c4d80c6c
15 changed files with 224 additions and 93 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
vllm/model_executor/models/terratorch.py
View File

@@ -64,7 +64,7 @@ from vllm.multimodal.profiling import BaseDummyInputsBuilder
from vllm.sequence import IntermediateTensors
from .interfaces import IsAttentionFree, MultiModalEmbeddings, SupportsMultiModal
from .interfaces_base import default_pooling_type
from .interfaces_base import attn_type
logger = init_logger(__name__)
@@ -220,7 +220,7 @@ class TerratorchMultiModalProcessor(BaseMultiModalProcessor):
)
@default_pooling_type("All")
@attn_type("attention_free")
@MULTIMODAL_REGISTRY.register_processor(
TerratorchMultiModalProcessor,
info=TerratorchProcessingInfo,