Skip to content

Pull requests: Lightning-AI/litgpt

New pull request New
16 Open 1,296 Closed
16 Open 1,296 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

chore(utils): add type annotations to public functions in utils.py
#2237 opened Apr 14, 2026 by nuthalapativarun Loading…
2
docs(pretrain): add TinyStories pretraining section
#2236 opened Apr 14, 2026 by nuthalapativarun Loading…
1
1
fix(model): convert bool mask_cache to float additive mask for softcapping
#2235 opened Apr 14, 2026 by nuthalapativarun Loading…
1
2
Multimodality support
#2232 opened Apr 9, 2026 by OnePunchMonk Contributor • Draft
fix(finetune): all_reduce val_loss for initial/final evaluation in multi-GPU training
#2224 opened Mar 30, 2026 by NIK-TIGER-BILL Loading…
3 tasks done
1
Fix division by zero in LR scheduler when max_steps equals warmup_steps
#2212 opened Mar 9, 2026 by Br1an67 Contributor Loading…
1
Fix Mistral tokenizer missing spaces in decode_stream (Issue #1822)
#2211 opened Mar 5, 2026 by mrshibly Loading…
1
1
DeepseekV3 models
#2175 opened Dec 29, 2025 by ysjprojects Collaborator Loading…
Fix IndexError in finetune scripts when last logit chunk becomes empty
#2141 opened Oct 4, 2025 by Copilot AI • Draft
@Borda
1
Added config for Mistral-Small-24B-Base-2501
#2136 opened Sep 25, 2025 by rcompt-lightning • Draft
@rcompt-lightning
1
Fix in LoRA code
#2114 opened Aug 28, 2025 by mseeger Contributor • Draft
@t-vi
9
Moving to lazy root imports to make config loading snappy enhancement New feature or request performance
#2073 opened Jun 13, 2025 by JackUrb Contributor • Draft
12
Add Multi-head Latent Attention (DeepSeekv2) enhancement New feature or request waiting on author
#1945 opened Feb 25, 2025 by simoneangarano • Draft
3
Raise error if disk is full before downloading weights bug Something isn't working waiting on author
#1903 opened Jan 8, 2025 by rasbt Contributor • Draft
1
7
OpenCoder series enhancement New feature or request waiting on author
#1880 opened Dec 21, 2024 by ysjprojects Collaborator • Draft
4
Add LongLora for both full and lora fine-tuning enhancement New feature or request waiting on author
#1350 opened Apr 24, 2024 by belerico Contributor • Draft
21
ProTip! Follow long discussions with comments:>50.