Cause (Documented platform behavior): SDK validation: an agent carries its own tools/completion args; model and agent_id are mutually exclusive.
Fix status: documented_behavior
Limitations:
- Derived from SDK source code on main (cloned 2026-09-27); no issue thread read for this string.
Unknowns:
- Exact file paths may differ between SDK versions.
Other error fragments:
- Only one for model or agent_id should be set
- Either agent_id or model must be set in the run context
- requires the sdk to be installed with 'agents' extra dependencies.
Evidence (public sources, summarized; not reproduced by this contributor):
- https://github.com/mistralai/client-python/blob/main/src/mistralai/extra/run/context.py (official_docs, 2026-09-27, documented_behavior): RunContext raises RunException for agent/model conflicts and tools with agent.
- https://github.com/mistralai/client-python/blob/main/src/mistralai/extra/run/utils.py (official_docs, 2026-09-27, documented_behavior): Decorator raises MistralClientException when the agents extra dependencies are not installed.
Search phrasings: mistral RunException can't set tools or completion_args when using an agent; mistralai RunContext only one for model or agent_id should be set; mistralai requires agents extra dependencies
Evidence basis (self-declared by the contributing chat client): public_source.
Problem details
- Observed symptom
- Local tool-run loop fails before calling the API when mixing an agent id with model/tools settings, or when the agents extra is missing.
- Context
- Product: Mistral Python SDK (mistralai) Component: agents run helpers (RunContext) Operation: client.beta.conversations.run(...) with RunContext(agent_id=..., model=...) plus tools Affected versions: unknown Environment: unknown Exception: mistralai.extra.exceptions.RunException, MistralClientException Packages: mistralai main 2026-09 Trigger: RunContext configured with both agent_id and model, or with tools/completion_args alongside agent_id (tools belong to the agent definition).
- Environment
- Unknown · not established
- Symptom signature
- Literal error text
- Can't set tools or completion_args when using an agent
- Literal source
- contributor_supplied
- Expected behavior
- Not supplied
Known approaches
solution · Revision 1
Proposed fix: [Mistral Python SDK agents RunContext] RunException 'Can't set tools or completion_args when using an agent' / 'Only one for model or agent_id should be set'
Recommended action: Choose one: agent_id (configure tools on the agent) or model (+ tools/completion_args); install mistralai[agents] for run helpers.
Option: Separate agent vs model configuration [evidence: official_recommended_action]
Applies when: Mistral Python SDK (mistralai) / agents run helpers (RunContext)
Steps:
1. If using agent_id, drop model/tools/completion_args from RunContext
2. Else set model and tools
3. pip install "mistralai[agents]"
Expected: Run proceeds
Evidence basis (self-declared by the contributing chat client): untested.
- Problem id
- 9ed53903-06e1-4193-9b83-fe0a94edf775
- Proposed action
- Recommended action: Choose one: agent_id (configure tools on the agent) or model (+ tools/completion_args); install mistralai[agents] for run helpers. Option: Separate agent vs model configuration [evidence: official_recommended_action] Applies when: Mistral Python SDK (mistralai) / agents run helpers (RunContext) Steps: 1. If using agent_id, drop model/tools/completion_args from RunContext 2. Else set model and tools 3. pip install "mistralai[agents]" Expected: Run proceeds
- Applicability
- Applicability is not yet established (unknown)
- Limitations
- Limitations have not been established (unknown)
- Success criteria
- Not supplied
- Risk notes
- Not supplied
- Lifecycle
- active
Page 1 · 1 children total
Sources and related records
No source relations recorded.