Boom! Ernie 4.5 MoE is working too on MLX! PR from Gökdeniz Gülmez is still a WIP but it works! Here ERNIE-4.5-21B-A3B-PT fp16 running on M3 Ultra 512GB. Generation ~10 t/s WIP Let's try The Beast ERNIE-4.5-300B-A47B-PT 🔥

Did Grok 4 just turn on? My ForYou feed just got a lot more interesting.

Vision LLM with MLX: Extracting Electric Meter Data in Production💡 I share my experience using the MLX backend to run Vision LLM (with MLX-VLM) for structured data extraction in a production environment. Thanks to MLX team Awni Hannun and Prince Canuma Try Sparrow:

Latest mlx-lm is out! pip install -U mlx-lm Bunch of new models: - SmolLM3 (Hugging Face) - Ernie family (Baidu) - BitNet (Microsoft) - Falcon-E (TII) - Text-only Gemma3n (Google) - MiniCPM4 (OpenBMB) - AFM (Apple) +Performance improvements for DWQ, dynamic quantization, and

Introducing Devstral Small and Medium 2507! This latest update offers improved performance and cost efficiency, perfectly suited for coding agents and software engineering tasks.

Detecting table columns for better structured data extraction 👀

Structured Data Query with Sparrow AI Agent Sparrow comes with option to extract stuctured data with query. In this video I explain how you can define such query to fetch array and field data. Video: youtube.com/watch?v=sGKvVX… GitHub: github.com/katanaml/sparr… Try Sparrow:

Thanks Prince. It was great to meet you in person! See you next time 💯

>>> Qwen3-Coder is here! ✅ We’re releasing Qwen3-Coder-480B-A35B-Instruct, our most powerful open agentic code model to date. This 480B-parameter Mixture-of-Experts model (35B active) natively supports 256K context and scales to 1M context with extrapolation. It achieves

Sparrow is getting good traction in the last few days 🚀 📟 Sparrow UI visits 🔝 🧭 Sparrow GitHub traffic 🔝 🤩 Sparrow GitHub stars 🔝 Sparrow UI: sparrow.katanaml.io Sparrow GitHub: github.com/katanaml/sparr…

The latest MLX has a CUDA back-end! To get started: pip install "mlx[cuda]" With the same codebase you can develop locally, run your model on Apple silicon, or in the cloud on Nvidia GPUs. MLX is designed around Apple silicon - which has a unified memory architecture. It uses

Meet the new Codestral 25.08 and the Complete Mistral Coding Stack for Enterprises! mistral.ai/news/codestral…

pip install -U mlx

Building new pipeline functionality in Sparrow github.com/katanaml/sparr… - AI assistant for code migration from Oracle Forms to APEX. Using Open Source LLMs, such as Mistral Codestral and Qwen Coder. This will run on local machine, completely private, no cloud dependency.

Gemma 3 270m 4-bit generates text at over 650 (!) tok/sec on an M4 Max with mlx-lm and uses < 200MB: Not sped up:

mlx-vlm v0.3.3 is here gr New models: - Liquid AI LFM2-VL - Z.ai GLM-4.5V - cohere Command-A-Vision Changes: - New kernel for grid_sample - Fix bicubic interpolate kernel compatibility with macOS < 15 - Fix config inheritance Thank you very much to all the amazing