Skip to content

Pull requests: InternLM/lmdeploy

New pull request New
59 Open 2,135 Closed
59 Open 2,135 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Support output_logits='generation' and output_last_hidden_state in PyTorch backend
#4534 opened Apr 17, 2026 by Copilot AI • Draft
1
@CUHKSZzxy
fix: prevent prefill starvation under high decode load
#4532 opened Apr 16, 2026 by grimoire Collaborator Loading…
7
Mixed modality
#4531 opened Apr 16, 2026 by CUHKSZzxy Collaborator Loading…
7
optimize get_sorted_idx in moe
#4529 opened Apr 15, 2026 by grimoire Collaborator Loading…
9
Test: update video sleep/wakeup and abort scenarios
#4528 opened Apr 15, 2026 by littlegy Contributor Loading…
1
style: add autopep8 pre-commit hook and apply PEP 8 formatting fixes
#4524 opened Apr 14, 2026 by windreamer Collaborator Loading…
3
Map user-input session_id to internal session_id to maintain session identity improvement
#4523 opened Apr 14, 2026 by lvhan028 Collaborator Loading…
11
[WIP]: Fix mtp experts
#4520 opened Apr 13, 2026 by RunningLeon Collaborator Loading…
2
fix qwen3.5 shared_expert_all_reduce
#4515 opened Apr 10, 2026 by yao-fengchen Collaborator • Draft
add explicit trust_remote_code controls to resolve the security issue improvement
#4511 opened Apr 8, 2026 by lvhan028 Collaborator Loading…
2
make fp8 model quantized by llm-compressor can be inferenced in turbomind enhancement New feature or request
#4509 opened Apr 8, 2026 by 43758726 Collaborator Loading…
8
support more message item types
#4501 opened Apr 7, 2026 by CUHKSZzxy Collaborator • Draft
fix: handle missing KV cache without crashing engine Bug:P0
#4497 opened Apr 4, 2026 by lvhan028 Collaborator Loading…
3
feat(turbomind): integrate cublasGemmGroupedBatchedEx for Qwen3.5 MoE inference on Blackwell GPUs with memory copy optimizations enhancement New feature or request
#4490 opened Apr 3, 2026 by hd9568 Loading…
23
Integrate deep-ep nccl backend enhancement New feature or request
#4477 opened Mar 27, 2026 by irexyc Collaborator Loading…
8
[refactor] [api_server] [1/N] Improve reasoning and tool-call parsers improvement
#4468 opened Mar 26, 2026 by lvhan028 Collaborator Loading…
8
feat: Turbomind linear gdn prefix caching enhancement New feature or request
#4465 opened Mar 25, 2026 by lapy Contributor Loading…
8
refactor get_ppl improvement
#4461 opened Mar 25, 2026 by lvhan028 Collaborator Loading…
8
feat: implement Turbomind vision encoder support for Qwen3VL/3.5 families enhancement New feature or request
#4460 opened Mar 24, 2026 by lapy Contributor Loading…
5
Support multi stop words improvement
#4454 opened Mar 24, 2026 by lvhan028 Collaborator Loading…
10
[Feature] Support n parameter in /v1/chat/completions and /v1/completions improvement
#4419 opened Mar 17, 2026 by ziyangliu-666 Loading…
1
22
[WIP] Support qwen3-omni
#4411 opened Mar 13, 2026 by CUHKSZzxy Collaborator • Draft
2 of 4 tasks
Add model deployment best practice section in user guide
#4399 opened Mar 9, 2026 by lvhan028 Collaborator • Draft
4
Fix Structured Output for GPT-OSS Models
#4386 opened Mar 2, 2026 by windreamer Collaborator Loading…
1
9
Improve proxy server improvement
#4354 opened Feb 12, 2026 by lvhan028 Collaborator Loading…
3
ProTip! What’s not been updated in a month: updated:<2026-03-17.