MLX
Fewer than 5 people recommend it so far, so there's no written verdict. Here is what commenters said.
4recommend
3warn against
Jan 15, 2026first mention
Oct 28, 2026latest mention
Why people recommend it
“I tried to tune it to use MLX on Mac GPUs, and it's so much faster.”
“I use LM Studio with MLX support, which is specifically faster on M series Macs.”
M series Mac useron “Bonsai 2 27B: Near-Lossless Compression in a 9x Smal…”
Show 2 more
“runs inference models fine, especially those using MLX”
“runs so well on Apple Silicon using MLX!”
Caveats and dissent
Show 1 more
“I tried 4bit MLX quants of Qwen 3.5 9B but it eventually would crash due to insufficient memory.”
Often compared with
Who recommends it in their own words
Threads where it came up
- Show HN: LFM3 Video Chat 1 mention
- Show HN: Use your OpenAI subscription to create audiobooks 1 mention
- Is Jev the general-purpose classifier we've been waiting for? 1 mention
- Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint 1 mention
- New Mac mini, featuring M6 and M5 Pro 1 mention
- llama.cpp 1 mention
Data: mlx.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 15
“I tried to tune it to use MLX on Mac GPUs, and it's so much faster.”
“using MLX and Kokoro”
“running on a consumer Apple laptop with MLX as the platform”
“I use LM Studio with MLX support, which is specifically faster on M series Macs.”
“runs inference models fine, especially those using MLX”
“mlx on the apple ASI (that never quite work out for me)”
“runs so well on Apple Silicon using MLX!”
“still faster than llama.cpp's 70.9t/s and MLX's 80.6t/s with the same model”
“MLX is not for end user deployment.”
“I tried 4bit MLX quants of Qwen 3.5 9B but it eventually would crash due to insufficient memory.”
“supports mlx for acceleration”
“port a diffusion language model to Swift+MLX so that I could run it on macOS”
“integration with MLX for hybrid GPU+ANE training”
“MLX tries to really leverage platform optimizations”
“only had to replace Transformers with Apple's MLX python library”