DeepSeek V4 Flash tokens per second
Live benchmark data for DeepSeek V4 Flash on opencode-go. Speed, latency, and reliability — benchmarked every ~10 minutes.
- Latest TPS
- 92.5 tok/s
- 24h avg TPS
- 89.8 tok/s
- TTFT
- 995ms
- 24h reliability
- 46%
- Last tested
- 8m 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 →