Repository navigation
CRITICAL: No effort is transmitted #34681
Description
Description
I'm using gpt-5.4 and opencode is not passing effort parameters
in opencode:
Proxy(no effort params):
in pi:
Proxy(effort max):
Plugins
no
OpenCode version
1.17.9+ea5dd9d
Steps to reproduce
- Setup config like this:
"bifrost-response": {
"models": {
"weegam/gpt-5.4": {
"limit": {
"context": 1047576,
"output": 65536
},
"name": "GPT-5.4",
"options": {
"include": [
"reasoning.encrypted_content"
],
"reasoningEffort": "high",
"reasoningSummary": "auto",
"textVerbosity": "low"
},
"reasoning": true,
"tool_call": true,
"variants": {
"high": {
"reasoningEffort": "high",
"reasoningSummary": "auto",
"textVerbosity": "low"
},
"low": {
"reasoningEffort": "low",
"reasoningSummary": "auto",
"textVerbosity": "low"
}
}
}
},
- Set effort to hight
- Give a request and see raw json
Screenshot and/or share link
No response
Operating System
NixOS Unstable
Terminal
Ghostty
Metadata
Metadata
Assignees
Labels
No labels
Activity
github-actions commentedon Jun 30, 2026
This issue might be a duplicate of existing issues. Please check:
options(reasoning/thinking) not forwarded to API for@ai-sdk/openai-compatibleproviders in headless mode #27361: Modeloptions(reasoning/thinking) not forwarded to API for@ai-sdk/openai-compatibleproviders in headless modeBoth report the same symptom — reasoning effort / model options not being transmitted to the upstream API. If your issue is distinct (e.g. it reproduces on a non-OpenAI-compatible provider path, or on a version where the above are fixed), please add that context so it can be triaged separately.
rekram1-node commentedon Jul 1, 2026
Ur prolly misconfiguring things cause it definitely works