{"model":{"id":"qwen3.5-9b-q8","name":"Qwen3.5 9B Instruct (Q8)","quantization":"Q8_0"},"hardware":{"slug":"rtx-5090","name":"NVIDIA GeForce RTX 5090","memoryGb":32},"verdict":"yes","localVerdict":"local_feasible","fitLevel":"Excellent","estimatedLoadGb":10.7,"usableMemoryGb":28.8,"estimatedTokensPerSec":81.3,"kvCacheByContext":[{"contextTokens":8192,"kvGb":0.25,"totalGb":10.95,"fits":true},{"contextTokens":16384,"kvGb":0.5,"totalGb":11.2,"fits":true},{"contextTokens":32768,"kvGb":1,"totalGb":11.7,"fits":true},{"contextTokens":65536,"kvGb":2,"totalGb":12.7,"fits":true},{"contextTokens":131072,"kvGb":4,"totalGb":14.7,"fits":true}],"upgradeGpu":null,"ollamaCommand":"ollama run qwen3.5:9b-q8_0","reportUrl":"https://modelfit.io/can-i-run/qwen3.5-9b-q8-on-rtx-5090/","source":"https://modelfit.io/","license":"CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/)","caveat":"Estimates from the ModelFit engine, not measurements."}