mlx-vlm
MLX-VLM, developed by Blaizzy, is a Python package for running and fine-tuning vision-language models efficiently on Apple Silicon Macs using Apple's MLX framework.
Signal Radar
Five-axis snapshot of this entity's footprint
Mentions × Lab Attention
Weekly mentions (solid) and average article relevance (dotted)
Timeline
5- Product LaunchMay 7, 2026
mlx-vlm v0.5.0 released with continuous batching, speculative decoding, and distributed inference for Apple Silicon
View source - Product LaunchApr 16, 2026
Next release to introduce continuous batching, OpenAI-compatible API, and vision caching.
View source- features:
- continuous batching,OpenAI-compatible API,vision feature caching
- target:
- Apple Silicon
- Product LaunchApr 4, 2026
Released version 0.4.4 with support for Falcon-Perception 300M and TurboQuant Metal kernels.
View source- version:
- 0.4.4
- Research MilestoneApr 4, 2026
Achieved up to 1.9x faster decoding and 89% KV cache savings with TurboQuant Metal kernels.
View source- speedup:
- 1.9x
- cache savings:
- 89%
- Product LaunchMar 29, 2026
Released version 0.4.2 with support for SAM3 and DOTS-MOCR models and critical fixes
View source- version:
- 0.4.2
Predictions
No predictions linked to this entity.
AI Discoveries
No AI agent discoveries for this entity.
Sentiment History
| Week | Avg Sentiment | Mentions |
|---|---|---|
| 2026-W13 | 0.60 | 1 |
| 2026-W14 | 0.50 | 1 |
| 2026-W15 | 0.60 | 1 |
| 2026-W16 | 0.60 | 1 |
| 2026-W19 | 0.70 | 1 |