feat(openrouter/deepseek/deepseek-v4-pro-0813): add new models [bot] - #2238
Conversation
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit 2d7f1b7. Configure here.
|
/test-models |
LordGameleo
left a comment
LordGameleo
left a comment
There was a problem hiding this comment.
Reviewed config only: single new model YAML under providers/openrouter/, schema-valid, costs/limits/modalities consistent with the linked source. Note: the Trigger gateway tests for changed models check is failing on this PR (not a required check).
…_tokens limits omitted max_output_tokens, so clients could treat generation as unbounded up to the 1M context window and send requests the API rejects. 384000 matches deepseek-v4-pro / v4-flash / v4-flash-0731.
|
/test-models |
LordGameleo
left a comment
LordGameleo
left a comment
There was a problem hiding this comment.
Bugbot finding verified as genuine against sibling configs and fixed in a follow-up commit; Bugbot has resolved the thread. Required YAML check green.
1 participant
Auto-generated by model-addition-agent for
openrouter/deepseek/deepseek-v4-pro-0813.Note
Low Risk
Catalog-only YAML addition for a new OpenRouter model; no runtime, auth, or data-handling logic is changed.
Overview
Adds serverless OpenRouter catalog entry for
deepseek/deepseek-v4-pro-0813.The model is chat-only, text in/out, with thinking enabled. It supports function calling, JSON/structured output, prompt caching, and tool choice. Limits are a 1,048,576 context window and 384,000 max output tokens, with default temperature and top_p of 1.
Reviewed by Cursor Bugbot for commit a1be518. Bugbot is set up for automated code reviews on this repo. Configure here.