Skip to content

Pull requests: InternLM/xtuner

New pull request New
61 Open 730 Closed
61 Open 730 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

add sft trainging with RL
#1403 opened Dec 30, 2025 by hhaAndroid Loading…
[Fix] grad_norm_dtype only support float32 and float64
#1402 opened Dec 30, 2025 by jayhenry Loading…
[CI] plot metric
#1401 opened Dec 30, 2025 by kkscilife Loading…
[Fix] Fix construct judger_cpu_pg in RL Trainer and add num_cpus_per_actor in NativeJudgerConfig
#1400 opened Dec 29, 2025 by YanhuiDua Loading…
[Fix] replace ray.get to await in async func and add timeout in ray.get
#1397 opened Dec 29, 2025 by YanhuiDua Loading…
8
Support tool call
#1395 opened Dec 25, 2025 by hhaAndroid Loading…
[Refactor] refactor packing in RL train controller and train worker
#1393 opened Dec 24, 2025 by YanhuiDua Loading…
26
support new model and fix some bugs
#1390 opened Dec 24, 2025 by RangiLyu Loading…
2
[Refactor] refactor trainer fit loop for better code organization
#1388 opened Dec 24, 2025 by HAOCHENYE Loading…
11
[Enhance] add training config printing to trainer
#1385 opened Dec 23, 2025 by HAOCHENYE Loading…
2
[Don't Merge] save dataset to save load time
#1379 opened Dec 19, 2025 by hhaAndroid Loading…
[CI]: add error-for-skips
#1376 opened Dec 18, 2025 by kkscilife Loading…
Update add_vision_id logic
#1372 opened Dec 17, 2025 by hhaAndroid Loading…
1
[Fix] Fix device mismatch for ascend npu when loss_cfg.loss_reduction != "token"
#1365 opened Dec 15, 2025 by nil0x9 Loading…
[Feature] support async rl
#1360 opened Dec 15, 2025 by YanhuiDua Loading…
4
[Draft] Add fsdp_mesh in load_spec
#1357 opened Dec 12, 2025 by HAOCHENYE Loading…
[Bug] Fix fp8 bugs: 1. support only fp8 quant gemm or grouped gemm 2. precompute float8 dynamic scale before compute_actor_logprobs during rl training
#1356 opened Dec 11, 2025 by HIT-cwh Loading…
[Debug] support using deemgemm bf16 gmm, do not merge
#1343 opened Dec 8, 2025 by HIT-cwh • Draft
[Feat][1/N] support async_rl in replaybuffer
#1337 opened Dec 5, 2025 by YanhuiDua Loading…
9
Fix small bug and support small features
#1324 opened Dec 3, 2025 by LKJacky Loading…
2
Draft: Support LoRA
#1304 opened Nov 27, 2025 by RangiLyu • Draft
4 of 7 tasks
[Enhance] Move pydantic model check to ut ci
#1297 opened Nov 25, 2025 by HAOCHENYE Loading…
3
[Fix] Fix Division-by-zero error when warmup steps set the same as total steps
#1285 opened Nov 21, 2025 by nil0x9 Loading…
[Fix] Add quantization config when saving fp8 and save tokenizer in offline dcp2hf
#1282 opened Nov 21, 2025 by HAOCHENYE Loading…
Support zero logprobs diff
#1273 opened Nov 18, 2025 by hhaAndroid Loading…
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.