Ollama v0.40.14 update released
The story so far
Ollama has released version 0.40.14, improving the output limits and API functions of the Claude Code model. The cloud model output of the Claude Code model is no longer limited to the default 128k. Meanwhile, the decoding issue of compressed search responses for the /api/chat endpoint has been fixed, ensuring the correct decoding of cloud search responses. Additionally, the OpenAI compatible endpoint no longer overrides the model's default sampling with wrapping. When the client does not send temperature/top_p parameters, the model's own values will be used. The /v1/responses endpoint now preserves readable input content, including previous assistant rejection messages. The Ollama application can also retry failed ChatGPT disconnections to avoid connection state corruption.
AI-written from the reports · Updated 16 days ago
Coverage timeline
Follow the coverage to see every side of the story.
- Ollama releasesv0.40.14
Ollama v0.40.14 released, Claude Code model output is no longer limited by the default 128k, cloud models can break this limit; fixed the decoding issue of compressed search responses for the /api/chat endpoint.