You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
During the free trail of Qwen 3.6 Plus, a context limit of 1M tokens was provided, which is the reason I chose Opencode Go, hoping to use it when the official paid access being available. But now I noticed that its context window has decreased to 260k in the new version of Opencode.
Is there any plan in future to bring back the old context window for Qwen3.6 Plus?
If your issue is specifically about the model's context window being reported incorrectly in OpenCode (rather than a Qwen API change), it likely overlaps with one of the above.
This seems to be an issue where the metadata on models.dev isn't setup correctly. This can be overwritten for now with a custom model config until they push the correction.
> This seems to be an issue where the metadata on models.dev isn't setup correctly. This can be overwritten for now with a custom model config until they push the correction.
>
> {
> "$schema": "https://opencode.ai/config.json",
> "provider": {
> "opencode-go": {
> "models": {
> "qwen3.6-plus": {
> "limit": {
> "context": 1000000,
> "output": 65536
> }
> }
> }
> }
> }
> }
Indeed, expliciltly declaring the context window limit bypassed the problem, which can also add some models supported by corresponding APIs but not in Opencode. Now we shall wait until further official notification.
Activity
github-actions commentedon Apr 15, 2026
This issue might be a duplicate of existing issues. Please check:
If your issue is specifically about the model's context window being reported incorrectly in OpenCode (rather than a Qwen API change), it likely overlaps with one of the above.
neuroconvergent commentedon Apr 17, 2026
This seems to be an issue where the metadata on models.dev isn't setup correctly. This can be overwritten for now with a custom model config until they push the correction.
{ "$schema": "https://opencode.ai/config.json", "provider": { "opencode-go": { "models": { "qwen3.6-plus": { "limit": { "context": 1000000, "output": 65536 } } } } } }Tao-Yida commentedon Apr 18, 2026
Indeed, expliciltly declaring the context window limit bypassed the problem, which can also add some models supported by corresponding APIs but not in Opencode. Now we shall wait until further official notification.
rekram1-node commentedon Jun 23, 2026
Automated: closing question issues.
GitHub Issues are for bug reports and feature requests only — questions / support are no longer tracked here.
If this is a bug or feature request, please open a new issue with the appropriate template.