Created bymario-alka.dePowered bynoob2claw.detricoma.de
⚡
Performance benchmark

How fast is the model?

Raw throughput on real hardware – how many tokens a model generates per second, how quickly it processes the prompt and how short the time to first token is. Higher is better.

⚡ Generation (tok/s)🔄 Prefill⏱️ TTFT👥 Concurrency
Reset
Metric:
Size
#Model / MakerMetricsParallelGPU / CPU / RAMRuntime
1Llama-3.2-1B-Instruct1BMeta Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
43,17 tok/s TG
Prefill 281 · TTFT 8.805 ms
1×AMD Liverpool (PlayStation 4)AMD Jaguar 8-Core (PlayStation 4) · 5 GB RAMllama.cppgodclawQ4_K_M
2Llama-3.2-1B-Instruct1BMeta Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
3,91 tok/s TG
Prefill 193 · TTFT 12.786 ms
1×Keine GPU (CPU-only)AMD Jaguar 8-Core (PlayStation 4) · 5 GB RAMllama.cppgodclawQ4_K_M
3Qwen3-0.6B0.6BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
3,80 tok/s TG
Prefill 23 · TTFT 106.805 ms
1×Keine GPU (CPU-only)AMD Jaguar 8-Core (PlayStation 4) · 5 GB RAMllama.cppgodclawQ4_K_M
4Qwen3-4B4BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
1,25 tok/s TG
Prefill 4 · TTFT 474.264 ms
1×Keine GPU (CPU-only)AMD Jaguar 8-Core (PlayStation 4) · 5 GB RAMllama.cppgodclawQ4_K_M