Ollama v0.40.15 Released with Support for MiMo V2.6 and AutoTrust
#Ollama #Alibaba Qwen #Anthropic #Apple
Ollama v0.40.15 has been released. It adds support for MiMo V2.6 (9B), built for code and agent workflows, and AutoTrust decision models, which return choices and probabilities from text.
Ollama v0.40.15 adds support for MiMo V2.6, a 9B model built for code and agent workflows. It also supports the AutoTrust decision models family, which includes 9b, 35b, and 27b-vl sizes; these models read text to return choices, yes/no probabilities, and ratings. The 35b and 27b-vl sizes can also read images, and in this release, autotrust-jev is supported on Apple silicon.
Other changes include ollama launch supporting Kimi Code CLI (the successor to kimi-cli) and offering to migrate existing kimi-cli setups. Additionally, ollama launch now automatically fills in Pi's model list, and Claude Desktop's model picker displays actual Ollama model names. Recurrent and vision GGUF models can now serve more than one request at a time, and Qwen3 and Qwen3.5 models use long-context (YaRN) scaling. Running ollama with no arguments now opens the welcome launcher.
This item is a summary. The full, checkable version is at the original source:
Read the original at Ollama ↗