🏆
Rankings
LLM Leaderboard
Real measurements of every tested language model on real, named hardware – rated by raw speed (performance in tokens per second, prefill and time to first token) and by practical task quality in complete agent and chat runs (harness). Pick a benchmark type below or filter by model, maker and hardware to see exactly what is tested and how the results are produced.
⚡ Performance (tok/s)🤖 Harness quality👥 Concurrency🖥️ real hardware
| # | Model / Maker | Metrics | Parallel | GPU / CPU / RAM | Runtime | ||
|---|---|---|---|---|---|---|---|
| 1 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 472,21 tok/s TG Prefill 4.673 · TTFT 9.026 ms | 10× | 2x NVIDIA RTX A6000 iAMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 2 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 447,71 tok/s TG Prefill 4.687 · TTFT 8.811 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 3 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 426,65 tok/s TG Prefill 4.869 · TTFT 8.761 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 4 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 424,22 tok/s TG Prefill 4.823 · TTFT 8.833 ms | 10× | 2x NVIDIA RTX A6000 iAMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 5 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 239,64 tok/s TG Prefill 4.683 · TTFT 2.060 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 6 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 238,93 tok/s TG Prefill 4.648 · TTFT 2.074 ms | 5× | 2x NVIDIA RTX A6000 iAMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 7 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 196,95 tok/s TG Prefill 4.698 · TTFT 4.796 ms | 10× | AMD Radeon AI PRO R9700AMD Ryzen 3 3100 4-Core Processor · 31 GB RAM | llama.cppopenclaw_cliQ8_0 | Details → | |
| 8 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 178,76 tok/s TG Prefill 4.812 · TTFT 2.004 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 9 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 178,14 tok/s TG Prefill 4.756 · TTFT 2.026 ms | 5× | 2x NVIDIA RTX A6000 iAMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 10 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 139,16 tok/s TG Prefill 3.596 · TTFT 2.796 ms | 5× | AMD Radeon AI PRO R9700AMD Ryzen 3 3100 4-Core Processor · 31 GB RAM | llama.cppopenclaw_cliQ8_0 | Details → | |
| 11 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 106,02 tok/s TG Prefill 3.586 · TTFT 538 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 12 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 105,97 tok/s TG Prefill 3.502 · TTFT 550 ms | 1× | 2x NVIDIA RTX A6000 iAMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 13 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 70,78 tok/s TG Prefill 3.529 · TTFT 546 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 14 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 70,63 tok/s TG Prefill 3.632 · TTFT 531 ms | 1× | 2x NVIDIA RTX A6000 iAMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 15 | Ornith-1.5-9B9BDeepReinforce Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 50,42 tok/s TG Prefill 2.788 · TTFT 691 ms | 1× | AMD Radeon AI PRO R9700AMD Ryzen 3 3100 4-Core Processor · 31 GB RAM | llama.cppopenclaw_cliQ8_0 | Details → |