AI Models

@cf/qwen/qwen2.5-coder-32b-instruct

Streams a reply to a ~20-word prompt, capped at 100 tokens; latency is the full reply.

3.21s?
Last Update35.35 s ago
Success (24h)100.0%
Predicted~3.41 s