Skip to content

[WIP] Interleave long-context prefill chunks with decode#4631

Draft
grimoire wants to merge 27 commits into
InternLM:mainfrom
grimoire:refactor-chunked-prefill
Draft

[WIP] Interleave long-context prefill chunks with decode#4631
grimoire wants to merge 27 commits into
InternLM:mainfrom
grimoire:refactor-chunked-prefill

Conversation

@grimoire

Copy link
Copy Markdown
Collaborator

requirement

Interleave chunk and decoding. Real prefix caching would be done in future PR.

@grimoire grimoire force-pushed the refactor-chunked-prefill branch from 2dc86e6 to 2775cd1 Compare June 1, 2026 11:37
grimoire added 2 commits June 2, 2026 16:11
…hing

# Conflicts:
#	lmdeploy/pytorch/multimodal/data_type.py
#	lmdeploy/vl/model/preprocess_utils.py
#	tests/test_lmdeploy/test_vl/test_preprocess_utils.py
@grimoire grimoire force-pushed the refactor-chunked-prefill branch from 80763a1 to 0f3284c Compare June 2, 2026 11:59
@grimoire grimoire force-pushed the refactor-chunked-prefill branch from 0f3284c to c76bae2 Compare June 8, 2026 04:11
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant