AI Models

@cf/meta/llama-3.3-70b-instruct-fp8-fast

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

2.15s?
Last Update14.35 s ago
Success (24h)100.0%
Predicted~2.15 s