Description
When using GPT-6 models through the GitHub Copilot provider, OpenCode does not include the selected reasoning effort in the outgoing /responses request.
OpenCode correctly records the selected variant, for example:
{
"providerID": "github-copilot",
"id": "gpt-6-luna",
"variant": "max"
}
but the outgoing request does not contain a reasoning field:
{
"model": "gpt-6-luna",
"include": ["reasoning.encrypted_content"],
"stream": true
}
The same happens with gpt-6-sol.
In comparison, gpt-5.6-luna with max correctly sends:
{
"model": "gpt-5.6-luna",
"reasoning": {
"effort": "max",
"summary": "auto"
},
"include": ["reasoning.encrypted_content"],
"stream": true
}
Plugins
OpenCode version
2.0.18
Steps to reproduce
No response
Screenshot and/or share link
No response
Operating System
No response
Terminal
No response
Description
When using GPT-6 models through the GitHub Copilot provider, OpenCode does not include the selected reasoning effort in the outgoing /responses request.
OpenCode correctly records the selected variant, for example:
{ "providerID": "github-copilot", "id": "gpt-6-luna", "variant": "max" }but the outgoing request does not contain a reasoning field:
{ "model": "gpt-6-luna", "include": ["reasoning.encrypted_content"], "stream": true }The same happens with gpt-6-sol.
In comparison, gpt-5.6-luna with max correctly sends:
{ "model": "gpt-5.6-luna", "reasoning": { "effort": "max", "summary": "auto" }, "include": ["reasoning.encrypted_content"], "stream": true }Plugins
OpenCode version
2.0.18
Steps to reproduce
No response
Screenshot and/or share link
No response
Operating System
No response
Terminal
No response