Qwen (Alibaba)
Qwen3-30B-A3B-Thinking-2507
Benchmarkprofil und veroeffentlichte Ergebnisse.
MoE30B
Einordnung im Feld
Bestwerte im Vergleich
Beste Messwerte dieses Modells gegen Minimum, Durchschnitt und Maximum aller veroeffentlichten Systeme.
Generation873,7 tok/s
Min 0,0Ø 89,4Max 1.295,3
Ueber dem Durchschnitt · 319 Systeme im Feld
Prefill33.023 tok/s
Min 6Ø 3.183Max 37.116
Ueber dem Durchschnitt · 318 Systeme im Feld
Time to First Token300 ms
Min 91Ø 21.684Max 299.591
Ueber dem Durchschnitt · 320 Systeme im Feld
Performanceprofil
Durchsatz nach Hardware & Engine
Jede Blase steht für eine GPU, CPU bzw. Engine – die Position zeigt Prompt-Verarbeitung (X) und Ausgabe-Geschwindigkeit (Y), die Blasengröße die Anzahl der Messläufe.
GPUnach Grafikkarte
NVIDIA GeForce RTX 5090 873,7 tok/sAMD Radeon AI PRO R9700 231,1 tok/sAMD Radeon 8060S Graphics 131,2 tok/sCPU-only 8,5 tok/s
CPUnach Prozessor
AMD Ryzen 7 5800X3D 8-Core Processor 873,7 tok/sAMD Ryzen Threadripper PRO 7955WX 16-Cores 231,1 tok/sAMD RYZEN AI MAX+ 395 w/ Radeon 8060S 131,2 tok/sIntel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz 8,5 tok/s
ENGnach Engine
vLLM 873,7 tok/sunbekannt 231,1 tok/sllama.cpp 91,4 tok/s
Agent- & Chat-Bewertung
Harnessbenchmark
Noch keine Harnessbenchmarks fuer dieses Modell.
Durchsatz & Latenz
Performancebenchmark
| # | Modell / Hersteller | Messwerte | Parallel | GPU / CPU / RAM | Runtime | ||
|---|---|---|---|---|---|---|---|
| 1 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 873,72 tok/s TG Prefill 33.023 · TTFT 1.835 ms | 10× | NVIDIA GeForce RTX 509016x AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | vLLMcodex_cliAWQ | Details → | |
| 2 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 633,41 tok/s TG Prefill 20.292 · TTFT 767 ms | 5× | NVIDIA GeForce RTX 509016x AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | vLLMcodex_cliAWQ | Details → | |
| 3 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 231,07 tok/s TG Prefill 3.322 · TTFT 3.312 ms | 5× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | godclawQ4_K_M | Details → | |
| 4 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 221,28 tok/s TG Prefill 8.379 · TTFT 300 ms | 1× | NVIDIA GeForce RTX 509016x AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | vLLMcodex_cliAWQ | Details → | |
| 5 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 197,45 tok/s TG Prefill 4.440 · TTFT 5.503 ms | 10× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | godclawQ4_K_M | Details → | |
| 6 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 131,22 tok/s TG Prefill 1.694 · TTFT 14.443 ms | 10× | AMD Radeon 8060S Graphics32x AMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | godclawQ4_K_M | Details → | |
| 7 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 122,51 tok/s TG Prefill 1.276 · TTFT 8.623 ms | 5× | AMD Radeon 8060S Graphics32x AMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | godclawQ4_K_M | Details → | |
| 8 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 91,39 tok/s TG Prefill 4.034 · TTFT 603 ms | 1× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 9 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 88,58 tok/s TG Prefill 4.831 · TTFT 503 ms | 1× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 10 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 87,84 tok/s TG Prefill 2.455 · TTFT 1.018 ms | 1× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 11 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 69,78 tok/s TG Prefill 1.153 · TTFT 1.880 ms | 1× | AMD Radeon 8060S Graphics32x AMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | godclawQ4_K_M | Details → | |
| 12 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 49,30 tok/s TG Prefill 320 · TTFT 6.725 ms | 1× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 13 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 48,40 tok/s TG Prefill 3.509 · TTFT 649 ms | 1× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 14 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 42,04 tok/s TG Prefill 3.904 · TTFT 581 ms | 1× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 15 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 41,95 tok/s TG Prefill 8.044 · TTFT 378 ms | 1× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 16 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 41,21 tok/s TG Prefill 5.446 · TTFT 460 ms | 1× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 17 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 8,52 tok/s TG Prefill 71 · TTFT 152.830 ms | 5× | Keine GPU (CPU-only)51x Intel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz · 504 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 18 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 6,64 tok/s TG Prefill 73 · TTFT 33.548 ms | 1× | Keine GPU (CPU-only)51x Intel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz · 504 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 19 | Qwen3-30B-A3B-Thinking-2507Qwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 0,58 tok/s TG Prefill 77 · TTFT 216.585 ms | 10× | Keine GPU (CPU-only)51x Intel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz · 504 GB RAM | llama.cppgodclawQ4_K_M | Details → |