🤖
Harness benchmark
How well does the model solve real tasks?
Quality on real agent and chat tasks – the model must handle real tasks with tools and multiple steps. Scored by success rate and task points achieved.
🎯 Success rate🏆 Task points🔧 Tool use🧠 real tasks
| # | Model / Maker | Metrics | GPU / CPU / RAM | Runtime | ||
|---|---|---|---|---|---|---|
| 1 | Kolibri-178.1BAleph Alpha Harness benchmarkOffice Agentic Work 100 - Tool Usage V1 | 100 Pkt 100,0% · 100/100 Aufg. · 547,7 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | officeagentic100NVFP4 | Details → | |
| 2 | Qwen3.6-35B-A3B-AWQ35.00BQwen (Alibaba) Harness benchmarkDecision-Benchmark V 1.0 | 200 Pkt 100,0% · 200/200 Aufg. · 57,0 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMBuzzwordTrainAWQ | Details → | |
| 3 | Qwen3.6-35B-A3B-AWQ35.00BQwen (Alibaba) Harness benchmarkDecision-Benchmark V 1.0 | 200 Pkt 100,0% · 200/200 Aufg. · 36,2 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMvllmAWQ | Details → | |
| 4 | Qwen3.8-27B-Uncensored27.00BQwen (Alibaba) Harness benchmarkDecision-Benchmark V 1.0 | 200 Pkt 100,0% · 200/200 Aufg. · 29,9 s | 2x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | vLLMvllm | Details → | |
| 5 | Open-Jev-27B-v1.127.00BZefan Cai / Open-Jev Harness benchmarkDecision-Benchmark V 1.0 | 200 Pkt 100,0% · 200/200 Aufg. · 75,5 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | pytorchpytorch | Details → | |
| 6 | Kolibri-178.1BAleph Alpha Harness benchmarkKolibri-Qwen Diverse Deterministic 210 V1.0 | 207 Pkt 98,6% · 207/210 Aufg. · 1.080,2 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMdeterministic210 | Details → | |
| 7 | Qwen3.6-35B-A3B-AWQ35.00BQwen (Alibaba) Harness benchmarkDecision-Benchmark High V 1.0 | 193 Pkt 96,5% · 193/200 Aufg. · 3.667,6 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMvllmAWQ | Details → | |
| 8 | Kolibri-178.1BAleph Alpha Harness benchmarkKolibri-Qwen Diverse Deterministic 310 V2.0 | 289 Pkt 93,2% · 289/310 Aufg. · 2.504,7 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMdeterministic310FP8 | Details → | |
| 9 | Kolibri-178.1BAleph Alpha Harness benchmarkKolibri-Qwen Diverse Deterministic 310 V2.0 | 283 Pkt 91,3% · 283/310 Aufg. · 2.384,4 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMdeterministic310NVFP4 | Details → | |
| 10 | 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 → | |
| 11 | MiniMax-M2.5230BMiniMax Harness benchmarkTool Usage Standard 1.0 | 2.427 Pkt 81,4% · 16/16 Aufg. · 228,0 s | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | vLLMgodclawBF16 | Details → | |
| 12 | Nemotron-Cascade-2-30B-A3B30BNVIDIA Harness benchmarkDecision-Benchmark V 1.0 | 159 Pkt 79,5% · 159/200 Aufg. · 35,0 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMvllm | Details → | |
| 13 | MiniMax-M2.5230BMiniMax Harness benchmarkTool Usage Standard 1.0 | 2.340 Pkt 78,5% · 16/16 Aufg. · 266,0 s | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | openclaw_cli | Details → | |
| 14 | MiniMax-M2.5230BMiniMax Harness benchmarkTool Usage Standard 1.0 | 2.033 Pkt 68,2% · 16/16 Aufg. · 373,0 s | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | vLLMgodclawBF16 | Details → | |
| 15 | MiniMax-M2.5230BMiniMax Harness benchmarkTools Usage Mini 1.0 | 130 Pkt 65,0% · 2/2 Aufg. · 70,0 s | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | godclaw | Details → | |
| 16 | LayaConvai Innovations Harness benchmarkDecision-Benchmark V 1.0 | 122 Pkt 61,0% · 122/200 Aufg. · 14,1 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | pytorchpytorch | Details → | |
| 17 | MiniMax-M2.5230BMiniMax Harness benchmarkTool-Parcours Dossier 40 V1.0 | 3.622 Pkt 60,8% · 34/40 Aufg. · 7.862,0 s | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | vLLMgodclawBF16 | Details → | |
| 18 | MiniMax-M2.5230BMiniMax Harness benchmarkTool Usage Standard 1.0 | 1.714 Pkt 57,5% · 16/16 Aufg. · 487,0 s | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | vLLMgodclawBF16 | Details → | |
| 19 | Qwen3.8-27B-Uncensored27.00BQwen (Alibaba) Harness benchmarkDecision-Benchmark High V 1.0 | 110 Pkt 55,0% · 110/200 Aufg. · 28,7 s | 2x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | vLLMvllm | Details → | |
| 20 | Qwen3.8-Flash-Next-Uncensored-NVFP4180BQwen (Alibaba) Harness benchmarkDecision-Benchmark High V 1.0 | 107 Pkt 53,5% · 107/200 Aufg. · 38,9 s | 2x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | vLLMvllmNVFP4 | Details → | |
| 21 | Kolibri-178.1BAleph Alpha Harness benchmarkKolibri-Qwen Diverse Deterministic 210 V1.0 | 108 Pkt 51,4% · 108/210 Aufg. · 72,1 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMdeterministic210 | Details → | |
| 22 | Qwen3.8-27B-Uncensored27.00BQwen (Alibaba) Harness benchmarkDecision-Benchmark High V 1.0 | 101 Pkt 50,5% · 101/200 Aufg. · 36,8 s | 2x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | vLLMvllm | Details → | |
| 23 | Qwen3.6-35B-A3B-AWQ35.00BQwen (Alibaba) Harness benchmarkDecision-Benchmark High V 1.0 | 87 Pkt 43,5% · 87/200 Aufg. · 19,0 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMvllmAWQ | Details → | |
| 24 | Qwen3.6-27B27BQwen (Alibaba) Harness benchmarkTool Usage Standard 1.0 | 1.220 Pkt 40,9% · 16/16 Aufg. · 1.036,0 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | vLLMgodclaw | Details → | |
| 25 | Kolibri-178.1BAleph Alpha Harness benchmarkKolibri-Qwen Diverse Deterministic 310 V2.0 | 119 Pkt 38,4% · 119/310 Aufg. · 87,9 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMdeterministic310NVFP4 | Details → | |
| 26 | Qwen3.6-27B27BQwen (Alibaba) Harness benchmarkTool-Parcours Dossier 40 V1.0 | 2.267 Pkt 38,0% · 38/40 Aufg. · 10.331,0 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | vLLMgodclaw | Details → | |
| 27 | Kolibri-178.1BAleph Alpha Harness benchmarkKolibri-Qwen Diverse Deterministic 310 V2.0 | 115 Pkt 37,1% · 115/310 Aufg. · 99,6 s | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMdeterministic310FP8 | Details → |

