Model components / Inference code

MLX-VLM

MLX-VLM runs vision-language and omni models locally on Apple silicon using MLX, taking image, audio and video inputs alongside text. It ships a CLI and a FastAPI server with continuous batching, automatic prefix caching, KV-cache quantization and speculative decoding, and converts and quantizes Hugging Face models for local use. Fine-tuning with LoRA and QLoRA is included but the package is inference-first.

Reached Round 1 as a finetuning_code candidate and was refiled here per ledger #431; the README is inference-first by a wide margin. Verified 2026-09-01 via GitHub, the LICENSE body, the repository README and the PyPI project JSON.

Openness

5 high confidence

5.0

license: MIT(OSI)
source: public
core-gated: ungated(no enterprise or paid tier in the repository or README

The LICENSE body is the stock MIT text, "Copyright (c) 2025 Prince Canuma", read in full rather than taken from the API label. The repository is public and unarchived and is the package itself. Nothing in the README or the repo offers a hosted, paid or enterprise build, so the core reads as ungated.

https://raw.githubusercontent.com/Blaizzy/mlx-vlm/main/LICENSE recorded 2026-09-01
LICENSE body, read in full: stock MIT text, "Copyright (c) 2025 Prince Canuma", no appended condition.
https://api.github.com/repos/Blaizzy/mlx-vlm recorded 2026-09-01
Repo metadata for Blaizzy/mlx-vlm - archived false, license MIT, 5,454 stars, pushed 2026-09-01.
https://raw.githubusercontent.com/Blaizzy/mlx-vlm/main/README.md recorded 2026-09-01
README, read for a paid tier or gated build (none found) and for the product's shape: "MLX-VLM is a package for inference and fine-tuning of Vision Language Models (VLMs) and Omni Models ... on your Mac using MLX"; FastAPI server, continuous batching, automatic prefix caching, KV cache quantization, speculative decoding, distributed inference; fine-tuning is one late section.

Adoption

3 high confidence

3.0

PyPI mlx-vlm drew 777,221 downloads in the trailing 30 days, banding at level 3 (100K-1M) on the software scale. The package backlinks to Blaizzy/mlx-vlm.

https://pypistats.org/api/packages/mlx-vlm/recent recorded 2026-09-01
"last_month":777221
https://pypi.org/pypi/mlx-vlm/json recorded 2026-09-01
project_urls Homepage/Repository = https://github.com/Blaizzy/mlx-vlm; version 0.6.17; summary names inference and fine-tuning of VLMs on Mac using MLX.

Capability

3 medium confidence

3.0

Two bands below the vllm anchor, level with uzu and litert-lm, the other single-target local engines: it serves one hardware family and does not carry the multi-node, multi-accelerator parallelism surface the bands above are placed on. Its multimodal input breadth (image, audio, video) is the widest of the local engines here.

https://raw.githubusercontent.com/Blaizzy/mlx-vlm/main/README.md recorded 2026-09-01
README sections: Server (FastAPI), Continuous Batching, Automatic Prefix Caching (APC), KV Cache Quantization, Speculative Decoding, Distributed Inference, Fine-tuning.

Verified 2026-09-01