{"verdict":"no","receipt":"https://fitllm.run/r/gpt-oss-120b-q4_k_m-on-rtx-4070","model":"gpt-oss-120b","hardware":"RTX 4070","env":"windows-display","quant":"Q4_K_M","kvBits":16,"ctx":8192,"memory":{"usedGB":77.11,"totalGB":12,"freeGB":-65.11,"breakdown":{"weightsGB":66.55,"kvCacheGB":0.29,"overheadGB":8.27,"reserveGB":2}},"maxContext":0,"contextLimits":{"recommendedMax":0,"hardLimit":0,"headroomPct":5},"fix":"A single A100 80GB would fit it.","note":"usedGB = total required to run: weights + KV cache + runtime overhead + OS/GPU reserve. Runtime resident-weights measurements (e.g. oMLX actual_size) read lower — a different metric, not a prediction error.","source":"https://fitllm.run","engine":"fitllm-engine (MIT)","data":"https://fitllm.run/data (CC0)"}