演示如何通过命令行在 Ollama 上运行新的 Qwen3.5-35B-A3B 模型进行编码任务。

编码实现📅 2026/03/31
#演示#开发者#GitHub#LLM 优化#低风险#手动触发#代码#代码仓库#手动#量化
Get started 

This preview release of Ollama accelerates the new Qwen3.5-35B-A3B model. 

Please make sure you have a Mac with more than 32GB of unified memory.

Claude Code:

ollama launch claude --model qwen3.5:35b-a3b-coding-nvfp4

OpenClaw:

ollama launch openclaw --model qwen3.5:35b-a3b-coding-nvfp4

Chat with the model:

ollama run qwen3.5:35b-a3b-coding-nvfp4

https://t.co/EaOnzdG9zt