Qwen3.8-Flash-Next speed on every GPU
2 measured runs across 1 GPUs, fastest first.
| GPU | Quant | Engine | tok/s | Roofline | Share |
|---|---|---|---|---|---|
| RTX 5090 | GSQ-RCO IQ1_M | llama.cpp + mods | 226.1 | 528.5 | 43% |
| RTX 5090 | GSQ-RCO IQ1_M | llama.cpp | 77.2 | 537.1 | 14% |
2 measured runs across 1 GPUs, fastest first.
| GPU | Quant | Engine | tok/s | Roofline | Share |
|---|---|---|---|---|---|
| RTX 5090 | GSQ-RCO IQ1_M | llama.cpp + mods | 226.1 | 528.5 | 43% |
| RTX 5090 | GSQ-RCO IQ1_M | llama.cpp | 77.2 | 537.1 | 14% |