Skip to content

Pull requests: karpathy/nanochat

New pull request New
34 Open 147 Closed
34 Open 147 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

remove unnecessary check to make the logic in CausalSelfAttention.forward() clearer refactor suggest/merge
#310 opened Nov 19, 2025 by ericsilberstein1 Loading…
1
make mid_train script work even with a tiny number of iterations improvement
#309 opened Nov 19, 2025 by ericsilberstein1 Loading…
rename checkpoint_dir to checkpoints_dir refactor suggest/merge
#308 opened Nov 19, 2025 by ericsilberstein1 Loading…
3
fixing two typos in comments docs Improvements or additions to documentation
#307 opened Nov 19, 2025 by ericsilberstein1 Loading…
change test/train split approach to fix bug in spelling bee task potential_bug Needs investigation/confirmation whether or not it's a bug
#306 opened Nov 19, 2025 by ericsilberstein1 Loading…
stats: include model_tag and model_step in /stats feature New feature or request
#296 opened Nov 16, 2025 by intently Loading…
Save checkpoint before possible OOM in CORE metric/Inference improvement
#295 opened Nov 16, 2025 by nitishpandey04 Loading…
Manage the Python module with maturin install
#294 opened Nov 14, 2025 by svlandeg • Draft
2
feat: Add TorchScript/ONNX export support for cross-language inference feature New feature or request
#275 opened Nov 10, 2025 by dhruvsoni365 Loading…
1
dataset downloader: add progress bar, essential info to stdout, and debug logs to logs/ (no functional change) feature New feature or request
#258 opened Nov 7, 2025 by h3nock Loading…
1
fix(common): avoid destroying non-initialized DDP process group potential_bug Needs investigation/confirmation whether or not it's a bug
#256 opened Nov 6, 2025 by dipeshbabu Loading…
1
Add centralized wandb initialization utility to reduce code duplication refactor
#253 opened Nov 5, 2025 by SermetPekin Loading…
2
Add pyproject for rustbpe standalone install
#252 opened Nov 5, 2025 by TensorTemplar Loading…
1
Compile eval model improvement suggest/review
#239 opened Nov 3, 2025 by SalmanMohammadi Loading…
QOL(shell scripts) install
#204 opened Oct 29, 2025 by kiankyars Loading…
feat(engine.py): Sample unique initial tokens for each sequence in a batch improvement
#201 opened Oct 29, 2025 by azekowka Loading…
1
FIX: uv workspace install todo
#172 opened Oct 25, 2025 by Majdoddin Loading…
21
Faster Regex pattern parsing in C feature New feature or request suggest/feedback
#161 opened Oct 23, 2025 by MadMax129 Loading…
2
Improve configurator: add testable parse_args() and ConfigManager class refactor
#159 opened Oct 23, 2025 by SermetPekin Loading…
Multi platform CI - Github Workflow tests v3 tests todo
#151 opened Oct 22, 2025 by SermetPekin Loading…
23
Fix mfu statically keyed to h100 max tflops improvement
#147 opened Oct 22, 2025 by Qubitium Loading…
1
Enabled markdown rendering for response by nanochat in ui.html todo
#141 opened Oct 21, 2025 by Goderr Loading…
5
add optional chunked cross-entropy for memory-efficient training feature New feature or request
#128 opened Oct 20, 2025 by mnehete32 Loading…
2
Enable Proxy/Gateway Compatibility via Uvicorn's --root-path improvement
#93 opened Oct 17, 2025 by wpybtw Loading…
4
Improve readability of huge numbers suggest/close UX
#90 opened Oct 16, 2025 by HosLak Loading…
1
ProTip! What’s not been updated in a month: updated:<2025-10-19.