{"model":{"id":"qwen3.5-4b-q4","name":"Qwen3.5 4B Instruct","quantization":"Q4_K_M"},"hardware":{"slug":"rtx-5070-ti","name":"NVIDIA GeForce RTX 5070 Ti","memoryGb":16},"verdict":"yes","localVerdict":"local_feasible","fitLevel":"Excellent","estimatedLoadGb":3.5,"usableMemoryGb":14.4,"estimatedTokensPerSec":156.8,"kvCacheByContext":[{"contextTokens":8192,"kvGb":0.25,"totalGb":3.75,"fits":true},{"contextTokens":16384,"kvGb":0.5,"totalGb":4,"fits":true},{"contextTokens":32768,"kvGb":1,"totalGb":4.5,"fits":true},{"contextTokens":65536,"kvGb":2,"totalGb":5.5,"fits":true},{"contextTokens":131072,"kvGb":4,"totalGb":7.5,"fits":true}],"upgradeGpu":null,"ollamaCommand":"ollama run qwen3.5:4b","reportUrl":"https://modelfit.io/can-i-run/qwen3.5-4b-q4-on-rtx-5070-ti/","source":"https://modelfit.io/","license":"CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/)","caveat":"Estimates from the ModelFit engine, not measurements."}