Owner

ml-explore

9 indexed repositories · View on GitHub

mlx
MLX: An array framework for Apple silicon

C++ · 28488 stars
mlx-examples
Examples in the MLX framework

Python · 8957 stars
mlx-lm
Run LLMs with MLX

Python · 7073 stars
mlx-swift-examples
Examples using MLX Swift

Swift · 2666 stars
mlx-swift
Swift API for MLX

C++ · 2037 stars
mlx-swift-lm
LLMs and VLMs with MLX Swift

Swift · 810 stars
mlx-data
Efficient framework-agnostic data loading

C++ · 483 stars
mlx-c
C API for MLX

C++ · 243 stars
mlx-onnx
MLX support for the Open Neural Network Exchange (ONNX)

72 stars

Beginner friendly Everything

Open All

feat: MXFP8 / fp8 mixed-precision training path (LoRA + full finetune) Open

Difficulty 5/5 Over a week Newbie friendliness 30/100

ml-explore/mlx-lm#1793 · Opened Aug 27, 2026
gated_delta: vectorized Metal kernel reads dt_bias per-head; callers (kimi_linear) pass per-element [H, head_dim] — GPU/CPU gate mismatch Open

Difficulty 2/5 1-3 hours Newbie friendliness 82/100

ml-explore/mlx-lm#1792 · Opened Aug 27, 2026
Parquet as a stream source: Python-level, or native alongside CSVReader/JSONLReader? Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

ml-explore/mlx-data#114 · 2 comments · Opened Aug 27, 2026
Proposal: neural surface-code decoder (QEC) example? Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

ml-explore/mlx-examples#1445 · Opened Aug 26, 2026
[BUG] NAX MoE simdgroup skipping in 0.32.2 triggers lost JACCL completion under heterogeneous TP Open

bug distributed

Difficulty 5/5 Over a week Newbie friendliness 38/100

ml-explore/mlx#4403 · 1 comment · Opened Aug 25, 2026
[BUG] Mixed slice + fancy indexing produces wrong shape/garbage data for out-of-range slice bounds Open

bug low priority

Difficulty 3/5 1-2 days Newbie friendliness 68/100

ml-explore/mlx#4399 · Opened Aug 25, 2026
Serving a sliding-window model alongside another model corrupts the other model's BatchKVCache: silent degenerate outputs on cache hits, then _idx=-1 crash (deterministic repro, 3/3; negative control 200 rounds clean) Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ml-explore/mlx-lm#1776 · Opened Aug 23, 2026
[BUG] mx.var / mx.std wrong for offset data Open

bug low priority

Difficulty 4/5 3-5 days Newbie friendliness 55/100

ml-explore/mlx#4379 · 3 comments · Opened Aug 23, 2026
mlx_median with no axis (axes=NULL, axes_num=0) fails: [flatten] start_axis must be less than or equal to end_axis Open

Difficulty 2/5 1-3 hours Newbie friendliness 84/100

ml-explore/mlx-c#128 · Opened Aug 23, 2026
MLX cannot find default.metallib under `swift test` with the Swift Build build system (aborts the whole test run) Open

Difficulty 3/5 1-2 days Newbie friendliness 82/100

ml-explore/mlx-swift#460 · 3 comments · Opened Aug 23, 2026
Suggestion: add a Stable Fast 3D (SF3D) example — native MLX single-image-to-3D-mesh Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ml-explore/mlx-examples#1443 · Opened Aug 22, 2026
Gemma 4 VLM path builds k_proj/v_proj for KV-shared layers, so E2B/E4B checkpoints fail to load Open

Difficulty 2/5 1-3 hours Newbie friendliness 88/100

ml-explore/mlx-swift-lm#558 · 3 comments · Opened Aug 20, 2026
[BUG] `__setitem__` with a single Ellipsis fails Open

bug low priority python

Difficulty 3/5 1-2 days Newbie friendliness 65/100

ml-explore/mlx#4358 · Opened Aug 19, 2026
Pipeline parallelism support for qwen3_moe Open

enhancement

Difficulty 3/5 1-2 days Newbie friendliness 72/100

ml-explore/mlx-lm#1763 · Opened Aug 19, 2026
Vendored mlx predates the June crash fix (a025496c): Metal command-buffer errors still hard-abort host apps Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

ml-explore/mlx-swift#458 · 3 comments · Opened Aug 19, 2026
Support fusing element-wise operations into reductions Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 45/100

ml-explore/mlx#4355 · 2 comments · Opened Aug 19, 2026
Core >= 0.31.2 thread-local streams break process-global Stream.gpu/.cpu — cross-thread eval aborts Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ml-explore/mlx-swift#457 · Opened Aug 18, 2026
[CUDA] gather_qmm ~24x slower than equal-FLOPs quantized_matmul on GB10 (sm_121) — makes quantized MoE prefill unusable Open

cuda performance

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ml-explore/mlx#4339 · 2 comments · Opened Aug 18, 2026
[whisper] `mlx_whisper` CLI overwrites output file when multiple audio inputs are provided Open

Difficulty 2/5 1-3 hours Newbie friendliness 85/100

ml-explore/mlx-examples#1441 · Opened Aug 18, 2026
whisper: numerical parity tests fail with MLX 0.32.0 on Apple Silicon Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ml-explore/mlx-examples#1440 · Opened Aug 18, 2026
iOS: supplying your own [KVCache] to generate() terminates the process with a larger model (macOS unaffected; MLX-managed cache unaffected) Open

Difficulty 4/5 3-5 days Newbie friendliness 45/100

ml-explore/mlx-swift-examples#492 · 1 comment · Opened Aug 17, 2026
vmap of scatter silently returns wrong values when the vmap axis is >= 2 Open

bug low priority

Difficulty 2/5 1-3 hours Newbie friendliness 72/100

ml-explore/mlx#4321 · 1 comment · Opened Aug 17, 2026
[BUG] JACCL SIGSEGV in tbt_post_recv during unique all_sum (no unplug, no model, ports still up) Open

bug distributed

Difficulty 4/5 3-5 days Newbie friendliness 38/100

ml-explore/mlx#4319 · 1 comment · Opened Aug 17, 2026
MLXFoundationModels does not compile against Xcode 27 beta 5 (27A5237l): LanguageModelCapabilities.init and metadata dictionaries changed Open

Difficulty 3/5 1-2 days Newbie friendliness 75/100

ml-explore/mlx-swift-lm#543 · 1 comment · Opened Aug 16, 2026
[BUG] JACCL never detects a lost peer: survivors busy-wait at 100% CPU indefinitely Open

bug distributed low priority

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ml-explore/mlx#4278 · 1 comment · Opened Aug 16, 2026
Ring backend: recv fails with errno 14 (EFAULT) during multi-node pipeline inference, fabric-independent (distinct from #3862) Open

bug distributed low priority

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ml-explore/mlx#4276 · 1 comment · Opened Aug 15, 2026
Distributed JACCL server crashes when two requests arrive concurrently Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

ml-explore/mlx-lm#1737 · 3 comments · Opened Aug 15, 2026
SDK-27 test bundle segfaults at startup on a macOS 26 host: weak-null FoundationModels metadata via TestResponseStream Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

ml-explore/mlx-swift-lm#539 · Opened Aug 15, 2026
Would new fused ops be welcome in mx.fast? Open

question

Difficulty 5/5 Over a week Newbie friendliness 48/100

ml-explore/mlx#4259 · 2 comments · Opened Aug 14, 2026
Code in the first example for Python SDK do not pass type check on load()'s returned values Open

Difficulty 2/5 1-3 hours Newbie friendliness 68/100

ml-explore/mlx-lm#1732 · Opened Aug 14, 2026
[BUG] JACCL segfaults in tbt_post_recv when a Thunderbolt cable is unplugged mid-collective Open

bug distributed

Difficulty 4/5 3-5 days Newbie friendliness 45/100

ml-explore/mlx#4192 · 1 comment · Opened Aug 11, 2026
Able to create MLXChatExampleApp, but the models do not respond. Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

ml-explore/mlx-swift-examples#490 · 1 comment · Opened Aug 11, 2026
ChatSession(instructions:) re-prefills the full system prompt on every turn — warm TTFT scales with instruction size on all tagged releases Open

Difficulty 3/5 1-2 days Newbie friendliness 58/100

ml-explore/mlx-swift-lm#522 · 2 comments · Opened Aug 10, 2026
Fused Metal KL-divergence kernel overflows (−3.39e38) in backward pass during DWQ for models >~1B Open

Difficulty 4/5 3-5 days Newbie friendliness 55/100

ml-explore/mlx-lm#1700 · 2 comments · Opened Aug 10, 2026
[BUG] `mx.distributed.all_max` / `all_min` silently drop `NaN` (inconsistent with `mx.max`/`mx.min`) Open

bug distributed low priority

Difficulty 3/5 1-2 days Newbie friendliness 68/100

ml-explore/mlx#4072 · 4 comments · Opened Aug 8, 2026
Idea: can we use swift-gigatoken? Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

ml-explore/mlx-swift-lm#517 · 4 comments · 1 reaction · Opened Aug 7, 2026
gemma3n: KV-shared layers only reuse cached KV when a cache is present, so no-cache forward (LoRA training, evaluation) diverges from generation Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

ml-explore/mlx-lm#1674 · Opened Aug 7, 2026
mlx_lm.generate: repetition_penalty has no default (unlike other sampling params), model can loop into repeated text with no warning Open

Difficulty 3/5 1-2 days Newbie friendliness 58/100

ml-explore/mlx-lm#1669 · 2 comments · Opened Aug 7, 2026
mlx_lm.fuse can not work Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

ml-explore/mlx-examples#1436 · Opened Aug 6, 2026
completeMemory() prepare count underflow kernel panic recurs under continuous batching despite #883's KV-cache-bound mitigations Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

ml-explore/mlx-lm#1666 · 2 comments · 2 reactions · Opened Aug 5, 2026
LoRA trainer has no gradient clipping or non-finite guard: one bad batch loses the whole run Open

Difficulty 4/5 3-5 days Newbie friendliness 55/100

ml-explore/mlx-lm#1658 · 1 comment · Opened Aug 5, 2026
chat prompt emits raw readline markers (\x01/\x02) because readline is never imported Open

Difficulty 3/5 1-2 days Newbie friendliness 72/100

ml-explore/mlx-lm#1653 · Opened Aug 3, 2026
load() pays an avoidable Hub round-trip per call even when the model is fully cached Open

Difficulty 2/5 1-3 hours Newbie friendliness 78/100

ml-explore/mlx-lm#1649 · Opened Aug 3, 2026
gemma3n_E2B hang and Metal-compiler jetsam under memory pressure Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ml-explore/mlx-swift-lm#498 · 1 comment · Opened Jul 31, 2026
olmoe_1B_7B coherence test disabled: GPTNeoXTokenizer unsupported by the tokenizer loader Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

ml-explore/mlx-swift-lm#497 · 1 comment · Opened Jul 31, 2026
31B MTP diagnostics never run on a cold CI runner (gated on cached ~35GB pair) Open

Difficulty 4/5 3-5 days Newbie friendliness 38/100

ml-explore/mlx-swift-lm#496 · 1 comment · Opened Jul 31, 2026
Verify opt-in env-var test gates propagate to the xctest process in CI Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

ml-explore/mlx-swift-lm#492 · Opened Jul 31, 2026
Support GLM-4-9B-0414 tool calling (markerless function-name + JSON) Open

Difficulty 4/5 3-5 days Newbie friendliness 68/100

ml-explore/mlx-swift-lm#491 · 1 comment · Opened Jul 31, 2026
[New feature] Support exporting memory snapshots compatible with PyTorch's memory viz format Open

enhancement low priority

Difficulty 5/5 Over a week Newbie friendliness 25/100

ml-explore/mlx#3954 · 2 comments · Opened Jul 31, 2026
Stable Diffusion: Transformer2D's GroupNorm uses MLX's default eps (1e-5) where diffusers uses 1e-6 Open

Difficulty 1/5 Under an hour Newbie friendliness 84/100

ml-explore/mlx-examples#1434 · Opened Jul 31, 2026

Showing the newest 100

This page lists what was indexed most recently. Advanced filter has the whole inventory, narrowed by language, difficulty and how long a task takes.

Open advanced filter

ml-explore

Showing the newest 100

Get new issues in your inbox