Qwen just published the 'Thinking' variant of this model 🔥 So you can run a model EVEN MORE powerful than GPT-4o locally! - Still only 3B active parameters - Open source license - 256k context window extendable to 1M - Strong in math, science and coding Details below ↓
Here're the perfect settings for this models in LM Studio: * Min P Sampling: 0 * Repeat Penalty: Disabled * Temperature: 0.6 * Top K Sampling: 20 * Top P Sampling: 0.95 You can set them in one click (+ the prompt template) here: https://lmstudio.ai/neil/qwen3-thinking

Useful links: - Hugging Face: https://huggingface.co/Qwen/Qwen3-30B-A3B-Thinking-2507 - Try it on Qwen Chat: http://chat.qwen.ai/?model=Qwen3-30B-A3B-2507 (click on "Thinking") One day, one open source model!
