Hi @proven basin. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
#Claude says it's 3.5 Sonnet.
pseudo axleBOT
Based on the knowledge sources, there is no explicit documentation about why Claude always reports itself as "3.5 Sonnet" in Warp, even when you have selected "3.7" or "4" as the model. However, there are a few relevant points:
- Warp allows you to select from multiple Anthropic models (including Claude 3.5 Sonnet, 3.7 Sonnet, and 4 Sonnet) for Agent Mode, and you can see and change the model in the dropdown above your Agent Mode prompts. Your selection should persist for future prompts, and the model being used is displayed in the UI. The documentation does not mention any known issue with the wrong model being used after selection, nor with the model misreporting its identity in responses Model Choice.
- There is no mention in the changelogs or blog posts of a bug where Claude always reports itself as "3.5 Sonnet" regardless of the actual model selected.
- Warp does use an "auto" mode that can select the best model for a given task, but when you manually select a model, it should use that model Model Choice.
Given this, it is possible that:
- The model is being selected correctly in Warp, but the Anthropic Claude models themselves may not always accurately report their own version in their responses.
- Alternatively, there could be a UI or backend issue where the selected model is not being used, but there is no documentation confirming this.
If you are consistently seeing this issue, it may be a bug or a limitation in how Claude identifies itself, or a mismatch between the selected model and the one actually used. The knowledge sources do not provide a definitive answer or workaround for this behavior.