[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"model-chronos-bolt-small":3},{"id":4,"slug":5,"name":6,"hf_id":6,"organization":7,"architecture":8,"architecture_config":9,"parameters":10,"parameters_label":11,"context_length":12,"license":12,"downloads":13,"likes":14,"status":15,"description":12,"huggingface_url":16,"created_at":17,"seo":18,"category":12,"family":21,"score":23,"variants":24,"quantizations":25,"requirements":52,"gpu_recommendations":110,"fitting_gpus":175,"community":211},466,"chronos-bolt-small","autogluon\u002Fchronos-bolt-small","autogluon","t5",[],47718016,"47.7M",null,9044420,61,"published","https:\u002F\u002Fhuggingface.co\u002Fautogluon\u002Fchronos-bolt-small","2026-08-13T18:04:35+00:00",{"title":6,"description":19,"image":12,"robots":20,"canonical_url":12},"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":7},"Autogluon",0.0149,[],[26,31,36,41,45,49],{"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":35},"FP16",16,1,0,{"format":37,"bits":38,"size_factor":39,"quality_loss":40},"FP8",8,0.5,0.01,{"format":42,"bits":28,"size_factor":43,"quality_loss":44},"GGUF",0.25,0.05,{"format":46,"bits":28,"size_factor":47,"quality_loss":48},"GPTQ",0.275,0.04,{"format":50,"bits":38,"size_factor":39,"quality_loss":51},"INT8",0.02,{"FP16":53,"FP8":76,"INT8":87,"AWQ":91,"GPTQ":102,"GGUF":106},{"minimum":54,"production":63,"recommended":70},{"scenario":55,"context_length":56,"batch_size":34,"weight_gb":57,"kv_cache_gb":39,"vram_gb":58,"ram_gb":59,"disk_gb":60,"requirement_source":61,"confidence":62},"minimum",4096,0.09,5.06,32,39.14,"inferred",0.95,{"scenario":64,"context_length":65,"batch_size":28,"weight_gb":57,"kv_cache_gb":66,"vram_gb":67,"ram_gb":68,"disk_gb":69,"requirement_source":61,"confidence":62},"production",16384,64,117.72,176.58,104.14,{"scenario":71,"context_length":72,"batch_size":73,"weight_gb":57,"kv_cache_gb":38,"vram_gb":74,"ram_gb":59,"disk_gb":75,"requirement_source":61,"confidence":62},"recommended",8192,2,20.82,65.14,{"minimum":77,"production":80,"recommended":84},{"scenario":55,"context_length":56,"batch_size":34,"weight_gb":48,"kv_cache_gb":39,"vram_gb":78,"ram_gb":59,"disk_gb":79,"requirement_source":61,"confidence":62},5.01,39.07,{"scenario":64,"context_length":65,"batch_size":28,"weight_gb":48,"kv_cache_gb":66,"vram_gb":81,"ram_gb":82,"disk_gb":83,"requirement_source":61,"confidence":62},117.66,176.49,104.07,{"scenario":71,"context_length":72,"batch_size":73,"weight_gb":48,"kv_cache_gb":38,"vram_gb":85,"ram_gb":59,"disk_gb":86,"requirement_source":61,"confidence":62},20.76,65.07,{"minimum":88,"production":89,"recommended":90},{"scenario":55,"context_length":56,"batch_size":34,"weight_gb":48,"kv_cache_gb":39,"vram_gb":78,"ram_gb":59,"disk_gb":79,"requirement_source":61,"confidence":62},{"scenario":64,"context_length":65,"batch_size":28,"weight_gb":48,"kv_cache_gb":66,"vram_gb":81,"ram_gb":82,"disk_gb":83,"requirement_source":61,"confidence":62},{"scenario":71,"context_length":72,"batch_size":73,"weight_gb":48,"kv_cache_gb":38,"vram_gb":85,"ram_gb":59,"disk_gb":86,"requirement_source":61,"confidence":62},{"minimum":92,"production":95,"recommended":99},{"scenario":55,"context_length":56,"batch_size":34,"weight_gb":51,"kv_cache_gb":39,"vram_gb":93,"ram_gb":59,"disk_gb":94,"requirement_source":61,"confidence":62},4.98,39.04,{"scenario":64,"context_length":65,"batch_size":28,"weight_gb":51,"kv_cache_gb":66,"vram_gb":96,"ram_gb":97,"disk_gb":98,"requirement_source":61,"confidence":62},117.63,176.45,104.04,{"scenario":71,"context_length":72,"batch_size":73,"weight_gb":51,"kv_cache_gb":38,"vram_gb":100,"ram_gb":59,"disk_gb":101,"requirement_source":61,"confidence":62},20.73,65.04,{"minimum":103,"production":104,"recommended":105},{"scenario":55,"context_length":56,"batch_size":34,"weight_gb":51,"kv_cache_gb":39,"vram_gb":93,"ram_gb":59,"disk_gb":94,"requirement_source":61,"confidence":62},{"scenario":64,"context_length":65,"batch_size":28,"weight_gb":51,"kv_cache_gb":66,"vram_gb":96,"ram_gb":97,"disk_gb":98,"requirement_source":61,"confidence":62},{"scenario":71,"context_length":72,"batch_size":73,"weight_gb":51,"kv_cache_gb":38,"vram_gb":100,"ram_gb":59,"disk_gb":101,"requirement_source":61,"confidence":62},{"minimum":107,"production":108,"recommended":109},{"scenario":55,"context_length":56,"batch_size":34,"weight_gb":51,"kv_cache_gb":39,"vram_gb":93,"ram_gb":59,"disk_gb":94,"requirement_source":61,"confidence":62},{"scenario":64,"context_length":65,"batch_size":28,"weight_gb":51,"kv_cache_gb":66,"vram_gb":96,"ram_gb":97,"disk_gb":98,"requirement_source":61,"confidence":62},{"scenario":71,"context_length":72,"batch_size":73,"weight_gb":51,"kv_cache_gb":38,"vram_gb":100,"ram_gb":59,"disk_gb":101,"requirement_source":61,"confidence":62},{"recommended":111,"cheapest":129,"performance":139,"min_complexity":156,"cluster_options":171,"no_match":172,"required_vram_gb":100,"candidate_count":173,"usage":174},{"gpu_model":112,"gpu_model_id":73,"gpu_count":34,"total_vram_gb":59,"hour_price":113,"monthly_cost":114,"providers":115,"bandwidth_gbps":118,"fp16_tflops":119,"interconnect_score":120,"parallel_efficiency":34,"score":121,"tier":122,"perf_metric":123,"perf_label":124,"perf_value":125,"workload":12,"cost_metric":126},"RTX 5090",0.3337,243.6,[116,117],"vast","runpod",1792,209,40,0.6705,"cheapest","tok\u002Fs","89600 tok\u002Fs",89600,{"effective_value":127,"effective_unit":123,"capacity_factor":128},62720,0.7,{"gpu_model":130,"gpu_model_id":131,"gpu_count":73,"total_vram_gb":59,"hour_price":132,"monthly_cost":133,"providers":134,"bandwidth_gbps":35,"fp16_tflops":35,"interconnect_score":120,"parallel_efficiency":135,"score":136,"tier":122,"perf_metric":123,"perf_label":137,"perf_value":35,"workload":12,"cost_metric":138},"Tesla V100",17,0.0289,42.19,[116],0.8,0.6194,"0 tok\u002Fs",{"effective_value":35,"effective_unit":123,"capacity_factor":128},{"gpu_model":140,"gpu_model_id":141,"gpu_count":38,"total_vram_gb":142,"hour_price":143,"monthly_cost":144,"providers":145,"bandwidth_gbps":147,"fp16_tflops":148,"interconnect_score":120,"parallel_efficiency":149,"score":150,"tier":151,"perf_metric":123,"perf_label":152,"perf_value":153,"workload":12,"cost_metric":154},"H200 SXM",11,1128,3.5,20440,[146],"lambda",4800,989,0.4,0.2197,"balanced","768000 tok\u002Fs",768000,{"effective_value":155,"effective_unit":123,"capacity_factor":128},537600,{"gpu_model":157,"gpu_model_id":158,"gpu_count":34,"total_vram_gb":159,"hour_price":160,"monthly_cost":161,"providers":162,"bandwidth_gbps":164,"fp16_tflops":165,"interconnect_score":120,"parallel_efficiency":34,"score":166,"tier":122,"perf_metric":123,"perf_label":167,"perf_value":168,"workload":12,"cost_metric":169},"RTX 3090",3,24,0.0622,45.41,[116,163],"tensordock",936,71,0.6654,"46800 tok\u002Fs",46800,{"effective_value":170,"effective_unit":123,"capacity_factor":128},32760,[],false,25,"value",[176,180,188,195,203],{"name":157,"vram_gb":159,"cheapest_hour":160,"cost":177,"provider":116,"estimated_tps":168},{"hourly":160,"daily":178,"monthly":161,"yearly":179},1.49,544.87,{"name":181,"vram_gb":159,"cheapest_hour":182,"cost":183,"provider":116,"estimated_tps":187},"RTX 4090",0.1356,{"hourly":182,"daily":184,"monthly":185,"yearly":186},3.25,98.99,1187.86,50400,{"name":189,"vram_gb":159,"cheapest_hour":190,"cost":191,"provider":116,"estimated_tps":187},"RTX 4090D",0.1602,{"hourly":190,"daily":192,"monthly":193,"yearly":194},3.84,116.95,1403.35,{"name":196,"vram_gb":159,"cheapest_hour":197,"cost":198,"provider":116,"estimated_tps":202},"A10",0.2015,{"hourly":197,"daily":199,"monthly":200,"yearly":201},4.84,147.1,1765.14,30000,{"name":204,"vram_gb":159,"cheapest_hour":205,"cost":206,"provider":116,"estimated_tps":210},"L4",0.2022,{"hourly":205,"daily":207,"monthly":208,"yearly":209},4.85,147.61,1771.27,15000,{"posts_count":35,"benchmarks_count":35}]