[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"model-devstral-small-2-24b":3},{"id":4,"slug":5,"name":6,"hf_id":7,"organization":8,"architecture":9,"architecture_config":8,"parameters":10,"parameters_label":11,"context_length":12,"license":13,"downloads":14,"likes":14,"status":15,"description":8,"huggingface_url":16,"created_at":17,"seo":18,"category":21,"family":8,"score":14,"variants":24,"quantizations":25,"requirements":51,"gpu_recommendations":134,"fitting_gpus":201,"community":236},112,"devstral-small-2-24b","Devstral Small 2 24B","mistralai\u002FDevstral-Small-2-24B",null,"transformer",24000000000,"24B",131072,"MIT",0,"published","https:\u002F\u002Fhuggingface.co\u002Fmistralai\u002FDevstral-Small-2-24B","2026-08-12T03:11:12+00:00",{"title":6,"description":19,"image":8,"robots":20,"canonical_url":8},"AI 模型部署决策引擎——开源 AI 的 GPU 需求、云端价格与成本分析。","index, follow",{"name":22,"slug":23},"Coding AI","coding",[],[26,31,35,40,44,48],{"format":27,"bits":28,"size_factor":29,"quality_loss":30},"AWQ",4,0.263,0.03,{"format":32,"bits":33,"size_factor":34,"quality_loss":14},"FP16",16,1,{"format":36,"bits":37,"size_factor":38,"quality_loss":39},"FP8",8,0.5,0.01,{"format":41,"bits":28,"size_factor":42,"quality_loss":43},"GGUF",0.25,0.05,{"format":45,"bits":28,"size_factor":46,"quality_loss":47},"GPTQ",0.275,0.04,{"format":49,"bits":37,"size_factor":38,"quality_loss":50},"INT8",0.02,{"FP16":52,"FP8":76,"INT8":90,"AWQ":94,"GPTQ":108,"GGUF":121},{"minimum":53,"production":62,"recommended":69},{"scenario":54,"context_length":55,"batch_size":34,"weight_gb":56,"kv_cache_gb":38,"vram_gb":57,"ram_gb":58,"disk_gb":59,"requirement_source":60,"confidence":61},"minimum",4096,44.7,61.5,92.25,108.74,"estimated",0.95,{"scenario":63,"context_length":64,"batch_size":28,"weight_gb":56,"kv_cache_gb":65,"vram_gb":66,"ram_gb":67,"disk_gb":68,"requirement_source":60,"confidence":61},"production",16384,64,179.29,268.94,173.74,{"scenario":70,"context_length":71,"batch_size":72,"weight_gb":56,"kv_cache_gb":37,"vram_gb":73,"ram_gb":74,"disk_gb":75,"requirement_source":60,"confidence":61},"recommended",8192,2,79.82,119.73,134.74,{"minimum":77,"production":82,"recommended":86},{"scenario":54,"context_length":55,"batch_size":34,"weight_gb":78,"kv_cache_gb":38,"vram_gb":79,"ram_gb":80,"disk_gb":81,"requirement_source":60,"confidence":61},22.35,33.22,49.84,73.87,{"scenario":63,"context_length":64,"batch_size":28,"weight_gb":78,"kv_cache_gb":65,"vram_gb":83,"ram_gb":84,"disk_gb":85,"requirement_source":60,"confidence":61},148.45,222.67,138.87,{"scenario":70,"context_length":71,"batch_size":72,"weight_gb":78,"kv_cache_gb":37,"vram_gb":87,"ram_gb":88,"disk_gb":89,"requirement_source":60,"confidence":61},50.26,75.39,99.87,{"minimum":91,"production":92,"recommended":93},{"scenario":54,"context_length":55,"batch_size":34,"weight_gb":78,"kv_cache_gb":38,"vram_gb":79,"ram_gb":80,"disk_gb":81,"requirement_source":60,"confidence":61},{"scenario":63,"context_length":64,"batch_size":28,"weight_gb":78,"kv_cache_gb":65,"vram_gb":83,"ram_gb":84,"disk_gb":85,"requirement_source":60,"confidence":61},{"scenario":70,"context_length":71,"batch_size":72,"weight_gb":78,"kv_cache_gb":37,"vram_gb":87,"ram_gb":88,"disk_gb":89,"requirement_source":60,"confidence":61},{"minimum":95,"production":100,"recommended":104},{"scenario":54,"context_length":55,"batch_size":34,"weight_gb":96,"kv_cache_gb":38,"vram_gb":97,"ram_gb":98,"disk_gb":99,"requirement_source":60,"confidence":61},11.59,19.62,32,57.09,{"scenario":63,"context_length":64,"batch_size":28,"weight_gb":96,"kv_cache_gb":65,"vram_gb":101,"ram_gb":102,"disk_gb":103,"requirement_source":60,"confidence":61},133.6,200.4,122.09,{"scenario":70,"context_length":71,"batch_size":72,"weight_gb":96,"kv_cache_gb":37,"vram_gb":105,"ram_gb":106,"disk_gb":107,"requirement_source":60,"confidence":61},36.03,54.05,83.09,{"minimum":109,"production":113,"recommended":117},{"scenario":54,"context_length":55,"batch_size":34,"weight_gb":110,"kv_cache_gb":38,"vram_gb":111,"ram_gb":98,"disk_gb":112,"requirement_source":60,"confidence":61},11.73,19.79,57.31,{"scenario":63,"context_length":64,"batch_size":28,"weight_gb":110,"kv_cache_gb":65,"vram_gb":114,"ram_gb":115,"disk_gb":116,"requirement_source":60,"confidence":61},133.79,200.69,122.31,{"scenario":70,"context_length":71,"batch_size":72,"weight_gb":110,"kv_cache_gb":37,"vram_gb":118,"ram_gb":119,"disk_gb":120,"requirement_source":60,"confidence":61},36.22,54.33,83.31,{"minimum":122,"production":126,"recommended":130},{"scenario":54,"context_length":55,"batch_size":34,"weight_gb":123,"kv_cache_gb":38,"vram_gb":124,"ram_gb":98,"disk_gb":125,"requirement_source":60,"confidence":61},12.01,20.15,57.74,{"scenario":63,"context_length":64,"batch_size":28,"weight_gb":123,"kv_cache_gb":65,"vram_gb":127,"ram_gb":128,"disk_gb":129,"requirement_source":60,"confidence":61},134.18,201.27,122.74,{"scenario":70,"context_length":71,"batch_size":72,"weight_gb":123,"kv_cache_gb":37,"vram_gb":131,"ram_gb":132,"disk_gb":133,"requirement_source":60,"confidence":61},36.59,54.88,83.74,{"recommended":135,"cheapest":157,"performance":168,"min_complexity":183,"cluster_options":197,"no_match":198,"required_vram_gb":105,"candidate_count":199,"usage":200},{"gpu_model":136,"gpu_model_id":137,"gpu_count":34,"total_vram_gb":138,"hour_price":139,"monthly_cost":140,"providers":141,"bandwidth_gbps":146,"fp16_tflops":147,"interconnect_score":148,"parallel_efficiency":34,"score":149,"tier":150,"perf_metric":151,"perf_label":152,"perf_value":153,"workload":8,"cost_metric":154},"H100 SXM",10,80,1.6022,1169.61,[142,143,144,145],"vast","tensordock","lambda","runpod",3350,66.9,40,0.6097,"balanced","tok\u002Fs","289 tok\u002Fs",289,{"effective_value":155,"effective_unit":151,"capacity_factor":156},202.3,0.7,{"gpu_model":158,"gpu_model_id":159,"gpu_count":28,"total_vram_gb":65,"hour_price":160,"monthly_cost":161,"providers":162,"bandwidth_gbps":14,"fp16_tflops":14,"interconnect_score":148,"parallel_efficiency":163,"score":164,"tier":165,"perf_metric":151,"perf_label":166,"perf_value":14,"workload":8,"cost_metric":167},"Tesla V100",17,0.0272,79.42,[142],0.6,0.5485,"cheapest","0 tok\u002Fs",{"effective_value":14,"effective_unit":151,"capacity_factor":156},{"gpu_model":169,"gpu_model_id":170,"gpu_count":37,"total_vram_gb":171,"hour_price":172,"monthly_cost":173,"providers":174,"bandwidth_gbps":175,"fp16_tflops":176,"interconnect_score":148,"parallel_efficiency":177,"score":178,"tier":150,"perf_metric":151,"perf_label":179,"perf_value":180,"workload":8,"cost_metric":181},"H200 SXM",11,1128,3.5,20440,[144],4800,989,0.4,0.2197,"1325.3 tok\u002Fs",1325.3,{"effective_value":182,"effective_unit":151,"capacity_factor":156},927.7,{"gpu_model":184,"gpu_model_id":185,"gpu_count":34,"total_vram_gb":186,"hour_price":187,"monthly_cost":188,"providers":189,"bandwidth_gbps":190,"fp16_tflops":191,"interconnect_score":148,"parallel_efficiency":34,"score":192,"tier":165,"perf_metric":151,"perf_label":193,"perf_value":194,"workload":8,"cost_metric":195},"RTX A6000",6,48,0.3001,219.07,[142,144],768,77,0.6908,"66.3 tok\u002Fs",66.3,{"effective_value":196,"effective_unit":151,"capacity_factor":156},46.4,[],false,23,"value",[202,206,214,221,228],{"name":184,"vram_gb":186,"cheapest_hour":187,"cost":203,"provider":142,"estimated_tps":194},{"hourly":187,"daily":204,"monthly":188,"yearly":205},7.2,2628.88,{"name":207,"vram_gb":186,"cheapest_hour":208,"cost":209,"provider":142,"estimated_tps":213},"L40S",0.4669,{"hourly":208,"daily":210,"monthly":211,"yearly":212},11.21,340.84,4090.04,74.5,{"name":215,"vram_gb":186,"cheapest_hour":216,"cost":217,"provider":142,"estimated_tps":14},"RTX PRO 6000 WS",0.8022,{"hourly":216,"daily":218,"monthly":219,"yearly":220},19.25,585.61,7027.27,{"name":222,"vram_gb":186,"cheapest_hour":223,"cost":224,"provider":142,"estimated_tps":14},"RTX PRO 6000 S",1.0131,{"hourly":223,"daily":225,"monthly":226,"yearly":227},24.31,739.56,8874.76,{"name":229,"vram_gb":138,"cheapest_hour":230,"cost":231,"provider":142,"estimated_tps":235},"A100 80GB",1.2,{"hourly":230,"daily":232,"monthly":233,"yearly":234},28.8,876,10512,175.9,{"posts_count":14,"benchmarks_count":14}]