Skip to content

Pull requests: lyogavin/airllm

New pull request New
39 Open 24 Closed
39 Open 24 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

feat: Add fixed-slot GPU expert streaming for PhiMoE
#350 opened Aug 24, 2026 by DevElCuy Loading…
feat: support prequantized BitsAndBytes streaming with Transformers 5
#347 opened Aug 23, 2026 by DevElCuy Loading…
Support nested Gemma 4 checkpoint layout
#346 opened Aug 22, 2026 by ousamabenyounes Loading…
1
Don't use per-expert streaming when shards are compressed
#344 opened Aug 21, 2026 by Hotragn Loading…
fix: robust layer-count parsing for nested checkpoint prefixes (complements #336)
#339 opened Aug 9, 2026 by HEETMEHTA18 Loading…
1
11
fix: handle missing files gracefully in remove_real_and_linked_file (#297)
#337 opened Aug 7, 2026 by bhumitschaudhry Loading…
1
6
docs: clarify compression performance tradeoffs
#334 opened Aug 6, 2026 by aryansk • Draft
1
3
fix: critical bugs — unsafe torch.load (RCE), UnboundLocalError crash, thread leak, broken tests
#333 opened Aug 5, 2026 by Sunil56224972 Loading…
docs: add Russian README translation
#332 opened Aug 5, 2026 by MonteNegroX Loading…
fix: avoid eager MLX LlamaForCausalLM import
#329 opened Aug 4, 2026 by maxitoon Loading…
1
docs: list Kimi K3 in Supported Models (#316)
#327 opened Aug 4, 2026 by evanclan Loading…
3 of 4 tasks
feat: AIRLLM_FULL_LOAD bypass — skip layer streaming when model fits in VRAM
#324 opened Aug 3, 2026 by he1074937561-a11y Loading…
3
Add DeepSeek V4 Flash selective expert streaming
#323 opened Aug 3, 2026 by francip Loading…
2
feat: add lightweight CPU-only CI/test harness
#322 opened Aug 2, 2026 by ardiannurcahya Loading…
1
Fix the unit tests so they can actually run
#315 opened Jul 28, 2026 by Hotragn Loading…
2
Use the model's real generation config for custom architectures
#313 opened Jul 24, 2026 by Hotragn Loading…
Fix 4-bit disk-space check demanding ~3.5x the model size
#309 opened Jul 23, 2026 by Hotragn Loading…
Default to trust_remote_code=False, fall back only when required (addresses #293)
#306 opened Jul 20, 2026 by Hotragn Loading…
1
Prefer safetensors over pickle .bin when splitting model layers (fixes #296)
#305 opened Jul 20, 2026 by Hotragn Loading…
1
Fix remove_real_and_linked_file double-deleting normal files
#303 opened Jul 19, 2026 by kapil971390 Loading…
1
1
Quality-of-life fork: refactored AutoModel, pyproject.toml, unit tests, CI
#291 opened Jun 4, 2026 by Doe-Johnevro Loading…
Fix per-layer streaming for transformers >=4.46
#288 opened May 21, 2026 by einyx Loading…
macOS MLX path: module reuse, mx.fast.* speedups, native 4bit
#286 opened May 8, 2026 by MartinLilt Loading…
3 of 4 tasks
Add Qwen3.6/Qwen3.5 support for newer Transformers versions
#283 opened May 1, 2026 by GanushevG Loading…
docs: setup project-specific open source standards and templates
#276 opened Apr 5, 2026 by uvenkatateja Loading…
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.