Skip to content

Pull requests: ml-explore/mlx-lm

New pull request New
219 Open 684 Closed
219 Open 684 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

Fix: qwen3_coder drops tool calls with float-formatted integer args
#1417 opened Jun 20, 2026 by brianmeyers007 Loading…
Add MiniMax-M3 tool-call parser
#1416 opened Jun 20, 2026 by anjansastry-hash Loading…
3 tasks done
Add Laguna M.1 model
#1415 opened Jun 18, 2026 by eauchs Loading…
fix: merge config.json eos_token_id with tokenizer-derived value instead of replacing it
#1414 opened Jun 18, 2026 by EmreCelenli Loading…
Fix gradient accumulation to weight micro-batches by token count
#1413 opened Jun 18, 2026 by Incheonkirin Loading…
Support per-layer DSA indexer schedule (GLM-5.2)
#1412 opened Jun 17, 2026 by davidrhodus Loading…
[transformers-to-mlx skill] glm_moe_dsa: DSA cross-layer indexer sharing (GLM-5.2)
#1410 opened Jun 16, 2026 by pcuenca • Draft
8
fix(server): reduce prompt cache memory pressure
#1409 opened Jun 16, 2026 by jsiu93 Loading…
1
fix(server): build xtc_special_tokens as a flat list
#1408 opened Jun 15, 2026 by devYRPauli Contributor Loading…
Add Quasar-Long (silx-ai/Quasar-Preview) model
#1407 opened Jun 15, 2026 by SahilChachra Contributor Loading…
fix(server): flush tool call when generation ends in "tool" state (Mistral/Devstral)
#1406 opened Jun 15, 2026 by rinaldofesta Loading…
server: persist LRU prompt cache to disk across restarts
#1405 opened Jun 15, 2026 by javruben Loading…
Add DiffusionGemma (text): block-diffusion model + diffusion_generate
#1402 opened Jun 14, 2026 by gilbert-barajas Loading…
4
Add MiniMax-M3 (text backbone)
#1401 opened Jun 13, 2026 by davidrhodus Loading…
1
feat(batch_generate): add return_logprobs for per-token log probabilities
#1400 opened Jun 12, 2026 by u7k4rs6 Loading…
4 tasks
1
fix: shorten generated tool-call ids to 9 alphanumeric chars (fixes #1375)
#1399 opened Jun 12, 2026 by u7k4rs6 Loading…
1
Add MiniMax M3 model support (text-only)
#1398 opened Jun 12, 2026 by machiabeli Loading…
6 tasks
2
feat: Add Qwen3.6 MoE (qwen3_5_moe) GGUF export support
#1397 opened Jun 12, 2026 by DeadByDawn101 Loading…
Fix Gemma 3n shared-KV layers with batched caches
#1396 opened Jun 11, 2026 by discobot Loading…
1
fix(mistral): parse JSON-list tool calls
#1394 opened Jun 11, 2026 by agisilaos Loading…
1
fix(server): honor prompt cache byte limit
#1392 opened Jun 11, 2026 by agisilaos Loading…
Add chunk-parallel gated delta ops for training
#1389 opened Jun 10, 2026 by tsato081 Loading…
2
Add CoPE (Clipped RoPE) rope type for zero-shot context extension
#1387 opened Jun 10, 2026 by machiabeli Loading…
2
Add gemma4_unified model alias for Gemma 4 12B support
#1386 opened Jun 10, 2026 by manzar-shahaan Loading…
feat: enhance chat CLI with readline history, line editing, and distributed support
#1382 opened Jun 9, 2026 by Vlor999 Loading…
1
11
ProTip! Adding no:label will show everything without a label.