Loading...
Loading...
Run side-by-side checks for pricing, context window, and latency.
giannisan
Open-source Laguna-S-2.1-Q2K-pulsar model from giannisan — available for download and self-hosting on Hugging Face.
vontra
Open-source Solar-Open2-250B-MLX-6bit model from vontra — available for download and self-hosting on Hugging Face.
atomicchat
Open-source ornith-35b-MLX-6bit model from atomicchat — available for download and self-hosting on Hugging Face.
Let our experts help you evaluate models based on your actual use case and budget. We'll provide a free analysis and actionable recommendations.
qwen
Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use. Built on the...