sullen echo
Hello. For the last ~30 minutes requests to https://openrouter.ai/anthropic/claude-3.7-sonnet model return reasoning block as part of their responses. It happens across at least Google and core Anthropic providers, so unlikely to be provider side issue.
lost kraken
That may happen if you're sending an optional reasoning parameter.
https://openrouter.ai/docs/use-cases/reasoning-tokens#controlling-reasoning-tokens
Or it still it still returning reasoning without it? Seems to work properly for me.