⚡
Performance-Benchmark
Wie schnell ist das Modell?
Roher Durchsatz auf echter Hardware – wie viele Token ein Modell pro Sekunde erzeugt, wie schnell es den Prompt verarbeitet und wie kurz die Zeit bis zum ersten Token ausfällt. Je höher, desto besser.
⚡ Generation (tok/s)🔄 Prefill⏱️ TTFT👥 Concurrency
| # | Modell / Hersteller | Messwerte | Parallel | GPU / CPU / RAM | Runtime | ||
|---|---|---|---|---|---|---|---|
| 1 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 388,05 tok/s TG Prefill 16.198 · TTFT 2.053 ms | 10× | 2x NVIDIA RTX A6000AMD EPYC 7203P 8-Core Processor · 252 GB RAM | vLLMgodclawINT4 | Details → | |
| 2 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 343,90 tok/s TG Prefill 9.629 · TTFT 1.973 ms | 5× | 2x NVIDIA RTX A6000AMD EPYC 7203P 8-Core Processor · 252 GB RAM | vLLMgodclawINT4 | Details → | |
| 3 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 99,60 tok/s TG Prefill 4.438 · TTFT 606 ms | 1× | 2x NVIDIA RTX A6000AMD EPYC 7203P 8-Core Processor · 252 GB RAM | vLLMgodclawINT4 | Details → | |
| 4 | Qwen3.8-Flash-Next180BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 71,73 tok/s TG Prefill 560 · TTFT 40.628 ms | 10× | 2x NVIDIA RTX A6000AMD EPYC 7203P 8-Core Processor · 252 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 5 | Qwen3.8-Flash-Next180BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 60,89 tok/s TG Prefill 409 · TTFT 24.788 ms | 5× | 2x NVIDIA RTX A6000AMD EPYC 7203P 8-Core Processor · 252 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 6 | Qwen3.8-Flash-Next180BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 36,99 tok/s TG Prefill 221 · TTFT 12.961 ms | 1× | 2x NVIDIA RTX A6000AMD EPYC 7203P 8-Core Processor · 252 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → |