What’s New in Ollama 0.40.15: MiMo V2.6, AutoTrust and Better Launch Integrations

Ollama 0.40.15 is out. (Copilot CLI 1.0.106 was already covered in 2719, so this post is Ollama only.)

• Xiaomi MiMo’s mimo-v2.6 (9B, built for code and agents): ollama run mimo-v2.6
• AutoTrust decision models (9b, 35b, 27b-vl) that return choices and probabilities for questions you supply; 35b and 27b-vl can read images. Apple silicon supported for autotrust-jev in this release
• ollama launch supports Kimi Code CLI (and can migrate kimi-cli setups), fills Pi’s model list, and shows real Ollama model names in Claude Desktop’s picker
• Recurrent and vision GGUF models can serve more than one request at a time
• Qwen3 / Qwen3.5 long-context (YaRN) scaling
• Running ollama with no arguments opens the welcome launcher

My take

Decision models that don’t free-generate are a nice tool for agent routing. Worth a try next to chat models.

Review before you install: read the release notes yourself, back up your config, and try new versions somewhere you can afford to break first.

God Bless, and Use the Source!

References & Further Reading

• Ollama v0.40.15 (GitHub)
• What’s New in Ollama 0.40.14 (EtherNinja)