Google
Gemma-4-26B-A4B-it26B
Benchmark profile and published results.
MoE26B
Position in the field
Best values compared
Best metrics of this model against the minimum, average and maximum of all published systems.
Generation1.589,1 tok/s
Min 0,0Ø 518,2Max 2.414,4
Ueber dem Durchschnitt · 244 Systeme im Feld
Prefill6.805 tok/s
Min 7Ø 6.972Max 39.029
Unter dem Durchschnitt · 244 Systeme im Feld
Time to First Token360 ms
Min 114Ø 27.182Max 354.674
Ueber dem Durchschnitt · 244 Systeme im Feld
Performance profile
Throughput by hardware & engine
Each bubble is a GPU, CPU or engine – position shows prompt processing (X) and output speed (Y), bubble size the number of runs.
GPUby graphics card
NVIDIA RTX PRO 6000 Blackwell Workstation Edition 1.589,1 tok/sNVIDIA GeForce RTX 5090 1.540,5 tok/sNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 1.479,9 tok/sNVIDIA GeForce RTX 5070 Ti 1.009,0 tok/sNVIDIA GeForce RTX 3090 Ti 926,2 tok/sNVIDIA RTX A6000 841,8 tok/sAMD Radeon PRO W7800 48GB 247,3 tok/sAMD Radeon 8060S Graphics 79,5 tok/s
CPUby processor
AMD Ryzen 9 9950X 16-Core Processor 1.589,1 tok/sAMD Ryzen 7 5800X3D 8-Core Processor 1.540,5 tok/sAMD Ryzen Threadripper PRO 9965WX 24-Cores 1.479,9 tok/sAMD Ryzen Threadripper PRO 5975WX 32-Cores 1.009,0 tok/sAMD Ryzen 9 8945HX with Radeon Graphics 926,2 tok/sAMD Ryzen Threadripper PRO 7955WX 16-Cores 841,8 tok/sAMD Ryzen 5 5600X 6-Core Processor 397,2 tok/sAMD RYZEN AI MAX+ 395 w/ Radeon 8060S 79,5 tok/s
Throughput & latency
Performance benchmark
| # | Model / Maker | Metrics | Parallel | GPU / CPU / RAM | Runtime | ||
|---|---|---|---|---|---|---|---|
| 1 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.589,11 tok/s TG Prefill 6.805 · TTFT 8.073 ms | 10× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 2 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.540,46 tok/s TG Prefill 4.499 · TTFT 10.048 ms | 10× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliUD-Q4_K_XL | Details → | |
| 3 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.479,91 tok/s TG Prefill 8.811 · TTFT 7.965 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 4 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.479,41 tok/s TG Prefill 6.414 · TTFT 8.654 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 5 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.466,18 tok/s TG Prefill 8.044 · TTFT 8.213 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 6 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.009,04 tok/s TG Prefill 4.090 · TTFT 12.734 ms | 10× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 7 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 926,20 tok/s TG Prefill 3.875 · TTFT 14.083 ms | 10× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketUD-Q4_K_XL | Details → | |
| 8 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 869,89 tok/s TG Prefill 6.141 · TTFT 2.760 ms | 5× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 9 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 841,75 tok/s TG Prefill 32.027 · TTFT 923 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 10 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 839,78 tok/s TG Prefill 4.050 · TTFT 3.627 ms | 5× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliUD-Q4_K_XL | Details → | |
| 11 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 800,90 tok/s TG Prefill 8.130 · TTFT 2.546 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 12 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 800,63 tok/s TG Prefill 7.264 · TTFT 2.674 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 13 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 796,50 tok/s TG Prefill 5.765 · TTFT 2.971 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 14 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 754,69 tok/s TG Prefill 26.306 · TTFT 1.120 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 15 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 690,88 tok/s TG Prefill 29.513 · TTFT 1.000 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawFP8 | Details → | |
| 16 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 630,94 tok/s TG Prefill 24.767 · TTFT 1.191 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawFP8 | Details → | |
| 17 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 585,92 tok/s TG Prefill 3.535 · TTFT 4.487 ms | 5× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 18 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 551,23 tok/s TG Prefill 24.679 · TTFT 522 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 19 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 522,97 tok/s TG Prefill 3.432 · TTFT 4.794 ms | 5× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketUD-Q4_K_XL | Details → | |
| 20 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 492,12 tok/s TG Prefill 20.612 · TTFT 628 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 21 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 488,90 tok/s TG Prefill 21.158 · TTFT 1.413 ms | 10× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMopenclaw_cliAWQ | Details → | |
| 22 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 459,02 tok/s TG Prefill 5.627 · TTFT 8.420 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 23 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 457,45 tok/s TG Prefill 5.564 · TTFT 8.488 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 24 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 447,47 tok/s TG Prefill 22.711 · TTFT 564 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawFP8 | Details → | |
| 25 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 408,21 tok/s TG Prefill 19.748 · TTFT 657 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawFP8 | Details → | |
| 26 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 406,14 tok/s TG Prefill 5.518 · TTFT 9.572 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 27 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 404,33 tok/s TG Prefill 5.768 · TTFT 9.559 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 28 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 397,18 tok/s TG Prefill 3.476 · TTFT 5.916 ms | 10× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 5 5600X 6-Core Processor · 30 GB RAM | llamacpp | Details → | |
| 29 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 393,62 tok/s TG Prefill 20.483 · TTFT 1.464 ms | 10× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMopenclaw_cliFP8 | Details → | |
| 30 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 321,71 tok/s TG Prefill 16.555 · TTFT 787 ms | 5× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMopenclaw_cliAWQ | Details → | |
| 31 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 287,13 tok/s TG Prefill 3.082 · TTFT 3.211 ms | 5× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 5 5600X 6-Core Processor · 30 GB RAM | llamacpp | Details → | |
| 32 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 272,45 tok/s TG Prefill 3.470 · TTFT 571 ms | 1× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliUD-Q4_K_XL | Details → | |
| 33 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 268,04 tok/s TG Prefill 5.498 · TTFT 360 ms | 1× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 34 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 263,67 tok/s TG Prefill 4.474 · TTFT 4.424 ms | 10× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 35 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 258,20 tok/s TG Prefill 16.062 · TTFT 813 ms | 5× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMopenclaw_cliFP8 | Details → | |
| 36 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 247,53 tok/s TG Prefill 5.604 · TTFT 353 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 37 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 247,30 tok/s TG Prefill 3.039 · TTFT 6.513 ms | 10× | 2x AMD Radeon PRO W7800 48GBAMD Ryzen Threadripper PRO 5975WX 32-Cores · 62 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 38 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 245,33 tok/s TG Prefill 7.549 · TTFT 262 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 39 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 245,29 tok/s TG Prefill 6.967 · TTFT 284 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 40 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 244,12 tok/s TG Prefill 2.809 · TTFT 7.040 ms | 10× | 2x AMD Radeon PRO W7800 48GBAMD Ryzen Threadripper PRO 5975WX 32-Cores · 62 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 41 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 228,28 tok/s TG Prefill 4.303 · TTFT 4.599 ms | 10× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ8_0 | Details → | |
| 42 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 222,45 tok/s TG Prefill 5.629 · TTFT 1.766 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 43 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 221,70 tok/s TG Prefill 5.542 · TTFT 1.796 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 44 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 200,22 tok/s TG Prefill 4.536 · TTFT 2.185 ms | 5× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 45 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 193,17 tok/s TG Prefill 5.827 · TTFT 1.699 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 46 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 191,88 tok/s TG Prefill 5.587 · TTFT 1.779 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 47 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 189,07 tok/s TG Prefill 3.110 · TTFT 638 ms | 1× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 48 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 175,20 tok/s TG Prefill 3.098 · TTFT 3.204 ms | 5× | 2x AMD Radeon PRO W7800 48GBAMD Ryzen Threadripper PRO 5975WX 32-Cores · 62 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 49 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 172,91 tok/s TG Prefill 2.730 · TTFT 3.618 ms | 5× | 2x AMD Radeon PRO W7800 48GBAMD Ryzen Threadripper PRO 5975WX 32-Cores · 62 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 50 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 169,27 tok/s TG Prefill 4.530 · TTFT 2.186 ms | 5× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ8_0 | Details → | |
| 51 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 166,99 tok/s TG Prefill 3.555 · TTFT 556 ms | 1× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketUD-Q4_K_XL | Details → | |
| 52 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 165,86 tok/s TG Prefill 10.425 · TTFT 190 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 53 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 156,19 tok/s TG Prefill 2.436 · TTFT 3.874 ms | 5× | AMD Radeon PRO W7800 48GBAMD Ryzen Threadripper PRO 5975WX 32-Cores · 62 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 54 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 153,31 tok/s TG Prefill 9.773 · TTFT 202 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawFP8 | Details → | |
| 55 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 145,21 tok/s TG Prefill 8.913 · TTFT 221 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 56 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 135,76 tok/s TG Prefill 8.302 · TTFT 238 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawFP8 | Details → | |
| 57 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 125,87 tok/s TG Prefill 3.206 · TTFT 618 ms | 1× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 5 5600X 6-Core Processor · 30 GB RAM | llamacpp | Details → | |
| 58 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 111,17 tok/s TG Prefill 4.594 · TTFT 430 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 59 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 111,16 tok/s TG Prefill 4.336 · TTFT 457 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 60 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 106,18 tok/s TG Prefill 7.369 · TTFT 268 ms | 1× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMopenclaw_cliAWQ | Details → | |
| 61 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 96,42 tok/s TG Prefill 6.483 · TTFT 305 ms | 1× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMopenclaw_cliFP8 | Details → | |
| 62 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 95,55 tok/s TG Prefill 4.144 · TTFT 479 ms | 1× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 63 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 94,54 tok/s TG Prefill 4.566 · TTFT 433 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 64 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 94,48 tok/s TG Prefill 4.490 · TTFT 440 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 65 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 87,26 tok/s TG Prefill 4.087 · TTFT 486 ms | 1× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ8_0 | Details → | |
| 66 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 79,45 tok/s TG Prefill 744 · TTFT 13.292 ms | 5× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 67 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 77,67 tok/s TG Prefill 745 · TTFT 13.316 ms | 5× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 68 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 76,33 tok/s TG Prefill 877 · TTFT 23.278 ms | 10× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 69 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 73,65 tok/s TG Prefill 843 · TTFT 24.099 ms | 10× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 70 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 68,19 tok/s TG Prefill 2.207 · TTFT 897 ms | 1× | AMD Radeon PRO W7800 48GBAMD Ryzen Threadripper PRO 5975WX 32-Cores · 62 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 71 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 65,98 tok/s TG Prefill 2.399 · TTFT 825 ms | 1× | 2x AMD Radeon PRO W7800 48GBAMD Ryzen Threadripper PRO 5975WX 32-Cores · 62 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 72 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 64,37 tok/s TG Prefill 2.232 · TTFT 886 ms | 1× | 2x AMD Radeon PRO W7800 48GBAMD Ryzen Threadripper PRO 5975WX 32-Cores · 62 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 73 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkPerformancetest Small 1.0 | 42,22 tok/s TG Prefill 71 · TTFT 609 ms | 1× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 74 | Gemma-4-26B-A4B-it26BGoogle Performance benchmark | 42,22 tok/s TG Prefill 71 · TTFT 609 ms | 1× | AMD Radeon 8060S Graphics32x AMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | godclawQ8_0 | Details → | |
| 75 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 39,53 tok/s TG Prefill 282 · TTFT 6.994 ms | 1× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 76 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,81 tok/s TG Prefill 499 · TTFT 4.595 ms | 1× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_0 | Details → | |
| 77 | Gemma-4-26B-A4B-it26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,39 tok/s TG Prefill 665 · TTFT 2.992 ms | 1× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_0 | Details → |
Agent & chat rating
Harness benchmark
| # | Model / Maker | Metrics | GPU / CPU / RAM | Runtime | ||
|---|---|---|---|---|---|---|
| 1 | Gemma-4-26B-A4B-it26BGoogle Harness benchmarkTool Usage Standard 1.0 | 2.701 Pkt 90,6% · 16/16 Aufg. · 118,0 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | openclaw_cli | Details → | |
| 2 | Gemma-4-26B-A4B-it26BGoogle Harness benchmarkTool Usage Standard 1.0 | 2.440 Pkt 81,9% · 16/16 Aufg. · 218,0 s | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 123 GB RAM | openclaw_cli | Details → | |
| 3 | Gemma-4-26B-A4B-it26BGoogle Harness benchmarkTool Usage Standard 1.0 | 2.383 Pkt 80,0% · 16/16 Aufg. · 245,0 s | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 123 GB RAM | openclaw_cli | Details → | |
| 4 | Gemma-4-26B-A4B-it26BGoogle Harness benchmarkTool Usage Standard 1.0 | 2.268 Pkt 76,1% · 16/16 Aufg. · 347,0 s | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 5 | Gemma-4-26B-A4B-it26BGoogle Harness benchmarkTool Usage Standard 1.0 | 2.197 Pkt 73,7% · 16/16 Aufg. · 375,0 s | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | openclaw_cli | Details → | |
| 6 | Gemma-4-26B-A4B-it26BGoogle Harness benchmarkGodBrain MCP Deep 40 - Wissensbasis-Lebenszyklus 1.0 | 2.271 Pkt 30,0% · 18/40 Aufg. · 982,0 s | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 7 | Gemma-4-26B-A4B-it26BGoogle Harness benchmarkTool-Parcours Dossier 40 V1.0 | 1.489 Pkt 25,0% · 22/40 Aufg. · 8.783,0 s | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ8_0 | Details → |