{"model":{"id":"gemma4-31b-q4","name":"Gemma 4 31B","quantization":"Q4_K_M"},"hardware":{"slug":"rtx-5080","name":"NVIDIA GeForce RTX 5080","memoryGb":16},"verdict":"tight","localVerdict":"local_slow","fitLevel":"Heavy","estimatedLoadGb":20,"usableMemoryGb":14.4,"estimatedTokensPerSec":5.9,"kvCacheByContext":[{"contextTokens":8192,"kvGb":2,"totalGb":22,"fits":false},{"contextTokens":16384,"kvGb":4,"totalGb":24,"fits":false},{"contextTokens":32768,"kvGb":8,"totalGb":28,"fits":false},{"contextTokens":65536,"kvGb":16,"totalGb":36,"fits":false},{"contextTokens":131072,"kvGb":32,"totalGb":52,"fits":false}],"upgradeGpu":{"slug":"rx-7900-xtx","name":"AMD Radeon RX 7900 XTX"},"ollamaCommand":"ollama run gemma4:31b","reportUrl":"https://modelfit.io/can-i-run/gemma4-31b-q4-on-rtx-5080/","source":"https://modelfit.io/","license":"CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/)","caveat":"Estimates from the ModelFit engine, not measurements."}