🏆
Ranglisten
LLM Leaderboard
Reale Messergebnisse aller getesteten Sprachmodelle auf echter, benannter Hardware – bewertet nach roher Geschwindigkeit (Performance in Token pro Sekunde, Prefill und Zeit bis zum ersten Token) und nach praktischer Aufgaben-Qualität in vollständigen Agenten- und Chat-Abläufen (Harness). Wähle unten einen Benchmark-Typ oder filtere nach Modell, Hersteller und Hardware, um genau zu sehen, was jeweils geprüft wird und wie die Ergebnisse zustande kommen.
⚡ Performance (tok/s)🤖 Harness-Qualität👥 Concurrency🖥️ echte Hardware
| # | Modell / Hersteller | Messwerte | Parallel | GPU / CPU / RAM | Runtime | ||
|---|---|---|---|---|---|---|---|
| 1 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.022,79 tok/s TG Prefill 18.744 · TTFT 2.204 ms | 10× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | vLLMgodclaw | Details → | |
| 2 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.022,79 tok/s TG Prefill 18.744 · TTFT 2.204 ms | 10× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | vLLMgodclaw | Details → | |
| 3 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 616,01 tok/s TG Prefill 13.778 · TTFT 803 ms | 5× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | vLLMgodclaw | Details → | |
| 4 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 616,01 tok/s TG Prefill 13.778 · TTFT 803 ms | 5× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | vLLMgodclaw | Details → | |
| 5 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 144,58 tok/s TG Prefill 7.078 · TTFT 281 ms | 1× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | vLLMgodclaw | Details → | |
| 6 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 144,58 tok/s TG Prefill 7.078 · TTFT 281 ms | 1× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | vLLMgodclaw | Details → | |
| 7 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 106,26 tok/s TG Prefill 3.145 · TTFT 6.467 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 8 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 102,29 tok/s TG Prefill 3.255 · TTFT 6.446 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 9 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 90,20 tok/s TG Prefill 3.429 · TTFT 5.889 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 10 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 90,13 tok/s TG Prefill 3.332 · TTFT 6.049 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 11 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 80,95 tok/s TG Prefill 2.992 · TTFT 8.497 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 12 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 79,73 tok/s TG Prefill 778 · TTFT 12.722 ms | 5× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 13 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 79,32 tok/s TG Prefill 763 · TTFT 12.964 ms | 5× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 14 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 78,87 tok/s TG Prefill 774 · TTFT 12.771 ms | 5× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 15 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 78,01 tok/s TG Prefill 753 · TTFT 13.128 ms | 5× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 16 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 77,67 tok/s TG Prefill 881 · TTFT 23.171 ms | 10× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 17 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 76,77 tok/s TG Prefill 886 · TTFT 23.041 ms | 10× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 18 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 76,77 tok/s TG Prefill 853 · TTFT 23.960 ms | 10× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 19 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 75,25 tok/s TG Prefill 878 · TTFT 23.266 ms | 10× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 20 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 59,13 tok/s TG Prefill 2.833 · TTFT 3.492 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 21 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 59,12 tok/s TG Prefill 2.866 · TTFT 3.452 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 22 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 58,64 tok/s TG Prefill 2.917 · TTFT 3.395 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 23 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 56,64 tok/s TG Prefill 3.217 · TTFT 6.338 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 24 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 55,17 tok/s TG Prefill 3.161 · TTFT 6.427 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 25 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 55,07 tok/s TG Prefill 3.075 · TTFT 6.594 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 26 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 45,96 tok/s TG Prefill 2.789 · TTFT 3.550 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 27 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 40,73 tok/s TG Prefill 2.786 · TTFT 3.598 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 28 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 39,82 tok/s TG Prefill 2.513 · TTFT 4.069 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 29 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,76 tok/s TG Prefill 781 · TTFT 2.535 ms | 1× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 30 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,62 tok/s TG Prefill 731 · TTFT 2.700 ms | 1× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 31 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,38 tok/s TG Prefill 757 · TTFT 2.611 ms | 1× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 32 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,36 tok/s TG Prefill 763 · TTFT 2.591 ms | 1× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 33 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,22 tok/s TG Prefill 2.641 · TTFT 3.757 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 34 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 22,25 tok/s TG Prefill 2.529 · TTFT 3.924 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 35 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 12,35 tok/s TG Prefill 1.938 · TTFT 1.019 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 36 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 12,35 tok/s TG Prefill 1.961 · TTFT 1.008 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 37 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 12,34 tok/s TG Prefill 1.960 · TTFT 1.007 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 38 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 12,33 tok/s TG Prefill 1.971 · TTFT 1.002 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 39 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 12,33 tok/s TG Prefill 1.960 · TTFT 1.008 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 40 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 12,33 tok/s TG Prefill 1.950 · TTFT 1.012 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 41 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 12,33 tok/s TG Prefill 1.951 · TTFT 1.012 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 42 | gemma-4-E4B-itGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 11,60 tok/s TG Prefill 1.763 · TTFT 1.142 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_0 | Details → |
