Qwen3 Coder 480B (non-reasoning) tokens per second
Live benchmark data for Qwen3 Coder 480B (non-reasoning) on ollama-free. Speed, latency, and reliability — benchmarked every ~10 minutes.
- Latest TPS
- 134.1 tok/s
- 24h avg TPS
- 134.1 tok/s
- TTFT
- 745ms
- 24h reliability
- 100%
- Last tested
- 4m ago
Tokens per second over time
Time to first token over time
Speed is measured as output tokens ÷ generation time. Each benchmark fires one streaming chat completion with a fixed ~300-token prompt, every ~10 minutes. Full methodology →