{"model":{"id":"lfm2-24b-a2b-q4","name":"LFM2 24B-A2B Instruct","quantization":"Q4_K_M"},"hardware":{"slug":"rtx-4070","name":"NVIDIA GeForce RTX 4070","memoryGb":12},"verdict":"tight","localVerdict":"local_slow","fitLevel":"Heavy","estimatedLoadGb":14,"usableMemoryGb":10.8,"estimatedTokensPerSec":44.1,"kvCacheByContext":[{"contextTokens":8192,"kvGb":2,"totalGb":16,"fits":false},{"contextTokens":16384,"kvGb":4,"totalGb":18,"fits":false},{"contextTokens":32768,"kvGb":8,"totalGb":22,"fits":false},{"contextTokens":65536,"kvGb":16,"totalGb":30,"fits":false},{"contextTokens":131072,"kvGb":32,"totalGb":46,"fits":false}],"upgradeGpu":{"slug":"rx-7900-xt","name":"AMD Radeon RX 7900 XT"},"ollamaCommand":"ollama run lfm2:24b-a2b","reportUrl":"https://modelfit.io/can-i-run/lfm2-24b-a2b-q4-on-rtx-4070/","source":"https://modelfit.io/","license":"CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/)","caveat":"Estimates from the ModelFit engine, not measurements."}