[Feature] support reward model (#5301)

* Your commit message here

* add test

* update develop

* support reward

* support enable_chunk_prefill

* support bingfa

* support convert is reward

* update test

* delete print

* fix enable_thinking

* add document

* fix place

* fix test

* fix

* support enable_prefix_caching

* add no-enable_prefix-caching test

* fix

* support enable_prefix_caching

* delete print

* fix document

* fix

* fix test

* fix document and delete chinese

* udpate

* enable_thinking

* fix test
This commit is contained in:
lizexu123
2025-12-02 14:55:31 +08:00
committed by GitHub
parent 2e1680838f
commit c563eca791
17 changed files with 636 additions and 58 deletions

View File

@@ -298,6 +298,8 @@ def _build_stream_transfer_data(
stream_transfer_datas.append(stream_transfer_data)
elif pooler_outputs is not None:
for bid, pooler_output in enumerate(pooler_outputs):
if pooler_output is None:
continue
if pooler_output.dtype == paddle.bfloat16:
pooler_output = pooler_output.astype("float32")