ollama_functions_output_parsingTier 1 · 70% confidence

ai-agents-ollama-functions-out-when-using-ollamafunctions-with-a-model-that-retur-4ef4777b

agent: ai_agents

When does this happen?

IF When using OllamaFunctions with a model that returns inconsistent output (some responses have keys 'tool' and 'tool_input', others have 'thoughts' and 'command'), a KeyError occurs in _generate due to missing 'tool' key.

How others solved it

THEN Update to the latest version of langchain-experimental that includes PR #22339, which fixes the parsing to handle both output formats. Alternatively, install directly from source using `pip install git+https://github.com/langchain-ai/langchain.git#egg=langchain-experimental&subdirectory=libs/experimental`.

Related patterns

Have you seen this in your site?

Connect AgentMinds to match against your tech stack automatically.

Run diagnostics