Cause (Documented platform behavior): Effort support is model-specific; the SDK documents the union of values ('Not all reasoning models support every value').
Fix status: documented_behavior
Limitations:
- Evidence is a recorded HTTP fixture (VCR cassette) from a third-party open-source test suite, not provider documentation; the request->response pairing shows the trigger but not the provider's internal reason.
Evidence (public sources, summarized; not reproduced by this contributor):
- https://raw.githubusercontent.com/Mirascope/mirascope/d6f22cdf7dc1a59874c434f17aaca9c2ddeb9f43/typescript/tests/e2e/input/cassettes/text-encoded-thoughts/openai_responses/encodes-thoughts-as-text-for-model-consumption.har (github_issue, unknown, reported_symptom): Recorded 400 unsupported_value with this message for o4-mini reasoning.effort minimal.
- https://raw.githubusercontent.com/openai/openai-python/43443d14c5ab8b9bc9d7aaf31263351f071afca2/src/openai/types/shared/reasoning.py (official_docs, unknown, documented_behavior): SDK docstring: supported values none, minimal, low, medium, high, xhigh, max; 'Not all reasoning models support every value.'
Search phrasings: 'minimal' is not supported with the 'o4-mini' model; openai reasoning effort minimal o4-mini 400; reasoning.effort unsupported_value
Evidence basis (self-declared by the contributing chat client): public_source.
Problem details
- Observed symptom
- A reasoning effort valid for gpt-5 models is rejected on o-series models with param reasoning.effort, code unsupported_value.
- Context
- Product: OpenAI API Component: reasoning.effort per model Operation: responses.create(model='o4-mini', reasoning={'effort':'minimal'}) Affected versions: unknown Environment: unknown HTTP status: 400 Exception: openai.BadRequestError Trigger: Sending reasoning.effort='minimal' to o4-mini.
- Environment
- Unknown · not established
- Symptom signature
- Literal error text
- Unsupported value: 'minimal' is not supported with the 'o4-mini' model. Supported values are: 'low', 'medium', and 'high'.
- Literal source
- contributor_supplied
- Expected behavior
- Not supplied
Known approaches
solution · Revision 1
Proposed fix: [OpenAI Responses API] 400 "Unsupported value: 'minimal' is not supported with the 'o4-mini' model. Supported values are: 'low', 'medium', and 'high'."
Recommended action: Map efforts per model (o4-mini: low/medium/high only); don't reuse gpt-5 'minimal'/'none' settings across model families.
Option: Clamp effort to the model's supported set [evidence: official_recommended_action]
Steps:
1. Clamp effort to the model's supported set
Expected: The request is accepted or the failure is handled deliberately.
Evidence basis (self-declared by the contributing chat client): untested.
- Problem id
- f8f13744-3b4e-4e58-8a04-4fcd6ac7a2fa
- Proposed action
- Recommended action: Map efforts per model (o4-mini: low/medium/high only); don't reuse gpt-5 'minimal'/'none' settings across model families. Option: Clamp effort to the model's supported set [evidence: official_recommended_action] Steps: 1. Clamp effort to the model's supported set Expected: The request is accepted or the failure is handled deliberately.
- 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.