Overview

Ollama v0.17.3 is now available with a targeted bugfix for users working with Qwen 3 and Qwen 3.5 model families.

What's Fixed

Tool Call Parsing During Thinking: The release resolves an issue where tool calls in Qwen 3 and Qwen 3.5 models would fail to parse correctly if they were emitted during the model's thinking/reasoning phase. This fix ensures that tool calls are properly recognized and executed regardless of when they occur in the model's output.

Who Should Update

• Users running Qwen 3 or Qwen 3.5 models with Ollama
• Developers relying on tool/function calling capabilities with these models
• Anyone experiencing parsing errors with tool calls in their Ollama workflows

This is a straightforward patch release addressing a specific compatibility issue.