[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"model-deepseek-r1-distill-qwen-15b":3},{"id":4,"slug":5,"name":6,"hf_id":7,"organization":8,"architecture":9,"architecture_config":10,"parameters":11,"parameters_label":12,"context_length":13,"license":14,"downloads":15,"likes":16,"status":17,"description":10,"huggingface_url":18,"created_at":19,"seo":20,"category":23,"family":26,"score":29,"variants":30,"quantizations":31,"requirements":58,"gpu_recommendations":139,"fitting_gpus":207,"community":243},102,"deepseek-r1-distill-qwen-15b","DeepSeek R1 Distill Qwen 1.5B","deepseek-ai\u002FDeepSeek-R1-Distill-Qwen-1.5B","deepseek-ai","transformer",null,1500000000,"1.5B",32768,"MIT",579747,1559,"published","https:\u002F\u002Fhuggingface.co\u002Fdeepseek-ai\u002FDeepSeek-R1-Distill-Qwen-1.5B","2026-08-12T03:11:11+00:00",{"title":6,"description":21,"image":10,"robots":22,"canonical_url":10},"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":24,"slug":25},"LLM","llm",{"name":27,"slug":28},"Qwen","qwen",0.0181,[],[32,37,42,47,51,55],{"format":33,"bits":34,"size_factor":35,"quality_loss":36},"AWQ",4,0.263,0.03,{"format":38,"bits":39,"size_factor":40,"quality_loss":41},"FP16",16,1,0,{"format":43,"bits":44,"size_factor":45,"quality_loss":46},"FP8",8,0.5,0.01,{"format":48,"bits":34,"size_factor":49,"quality_loss":50},"GGUF",0.25,0.05,{"format":52,"bits":34,"size_factor":53,"quality_loss":54},"GPTQ",0.275,0.04,{"format":56,"bits":44,"size_factor":45,"quality_loss":57},"INT8",0.02,{"FP16":59,"FP8":83,"INT8":96,"AWQ":100,"GPTQ":113,"GGUF":126},{"minimum":60,"production":69,"recommended":76},{"scenario":61,"context_length":62,"batch_size":40,"weight_gb":63,"kv_cache_gb":45,"vram_gb":64,"ram_gb":65,"disk_gb":66,"requirement_source":67,"confidence":68},"minimum",4096,2.79,8.48,32,43.36,"estimated",0.95,{"scenario":70,"context_length":71,"batch_size":34,"weight_gb":63,"kv_cache_gb":72,"vram_gb":73,"ram_gb":74,"disk_gb":75,"requirement_source":67,"confidence":68},"production",16384,64,121.46,182.18,108.36,{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":63,"kv_cache_gb":44,"vram_gb":80,"ram_gb":81,"disk_gb":82,"requirement_source":67,"confidence":68},"recommended",8192,2,24.4,36.59,69.36,{"minimum":84,"production":88,"recommended":92},{"scenario":61,"context_length":62,"batch_size":40,"weight_gb":85,"kv_cache_gb":45,"vram_gb":86,"ram_gb":65,"disk_gb":87,"requirement_source":67,"confidence":68},1.4,6.72,41.18,{"scenario":70,"context_length":71,"batch_size":34,"weight_gb":85,"kv_cache_gb":72,"vram_gb":89,"ram_gb":90,"disk_gb":91,"requirement_source":67,"confidence":68},119.53,179.29,106.18,{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":85,"kv_cache_gb":44,"vram_gb":93,"ram_gb":94,"disk_gb":95,"requirement_source":67,"confidence":68},22.55,33.82,67.18,{"minimum":97,"production":98,"recommended":99},{"scenario":61,"context_length":62,"batch_size":40,"weight_gb":85,"kv_cache_gb":45,"vram_gb":86,"ram_gb":65,"disk_gb":87,"requirement_source":67,"confidence":68},{"scenario":70,"context_length":71,"batch_size":34,"weight_gb":85,"kv_cache_gb":72,"vram_gb":89,"ram_gb":90,"disk_gb":91,"requirement_source":67,"confidence":68},{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":85,"kv_cache_gb":44,"vram_gb":93,"ram_gb":94,"disk_gb":95,"requirement_source":67,"confidence":68},{"minimum":101,"production":105,"recommended":109},{"scenario":61,"context_length":62,"batch_size":40,"weight_gb":102,"kv_cache_gb":45,"vram_gb":103,"ram_gb":65,"disk_gb":104,"requirement_source":67,"confidence":68},0.72,5.87,40.13,{"scenario":70,"context_length":71,"batch_size":34,"weight_gb":102,"kv_cache_gb":72,"vram_gb":106,"ram_gb":107,"disk_gb":108,"requirement_source":67,"confidence":68},118.6,177.9,105.13,{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":102,"kv_cache_gb":44,"vram_gb":110,"ram_gb":111,"disk_gb":112,"requirement_source":67,"confidence":68},21.66,32.49,66.13,{"minimum":114,"production":118,"recommended":122},{"scenario":61,"context_length":62,"batch_size":40,"weight_gb":115,"kv_cache_gb":45,"vram_gb":116,"ram_gb":65,"disk_gb":117,"requirement_source":67,"confidence":68},0.73,5.88,40.14,{"scenario":70,"context_length":71,"batch_size":34,"weight_gb":115,"kv_cache_gb":72,"vram_gb":119,"ram_gb":120,"disk_gb":121,"requirement_source":67,"confidence":68},118.61,177.92,105.14,{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":115,"kv_cache_gb":44,"vram_gb":123,"ram_gb":124,"disk_gb":125,"requirement_source":67,"confidence":68},21.67,32.5,66.14,{"minimum":127,"production":131,"recommended":135},{"scenario":61,"context_length":62,"batch_size":40,"weight_gb":128,"kv_cache_gb":45,"vram_gb":129,"ram_gb":65,"disk_gb":130,"requirement_source":67,"confidence":68},0.75,5.9,40.17,{"scenario":70,"context_length":71,"batch_size":34,"weight_gb":128,"kv_cache_gb":72,"vram_gb":132,"ram_gb":133,"disk_gb":134,"requirement_source":67,"confidence":68},118.64,177.95,105.17,{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":128,"kv_cache_gb":44,"vram_gb":136,"ram_gb":137,"disk_gb":138,"requirement_source":67,"confidence":68},21.69,32.54,66.17,{"recommended":140,"cheapest":158,"performance":173,"min_complexity":190,"cluster_options":204,"no_match":205,"required_vram_gb":110,"candidate_count":161,"usage":206},{"gpu_model":141,"gpu_model_id":79,"gpu_count":40,"total_vram_gb":65,"hour_price":142,"monthly_cost":143,"providers":144,"bandwidth_gbps":147,"fp16_tflops":148,"interconnect_score":149,"parallel_efficiency":40,"score":150,"tier":151,"perf_metric":152,"perf_label":153,"perf_value":154,"workload":10,"cost_metric":155},"RTX 5090",0.337,246.01,[145,146],"vast","runpod",1792,209,40,0.6782,"cheapest","tok\u002Fs","2488.9 tok\u002Fs",2488.9,{"effective_value":156,"effective_unit":152,"capacity_factor":157},1742.2,0.7,{"gpu_model":159,"gpu_model_id":160,"gpu_count":79,"total_vram_gb":161,"hour_price":162,"monthly_cost":163,"providers":164,"bandwidth_gbps":165,"fp16_tflops":166,"interconnect_score":149,"parallel_efficiency":167,"score":168,"tier":151,"perf_metric":152,"perf_label":169,"perf_value":170,"workload":10,"cost_metric":171},"RTX 5070",28,24,0.0101,14.75,[145],672,61,0.8,0.6142,"1493.3 tok\u002Fs",1493.3,{"effective_value":172,"effective_unit":152,"capacity_factor":157},1045.3,{"gpu_model":174,"gpu_model_id":175,"gpu_count":44,"total_vram_gb":176,"hour_price":177,"monthly_cost":178,"providers":179,"bandwidth_gbps":181,"fp16_tflops":182,"interconnect_score":149,"parallel_efficiency":183,"score":184,"tier":185,"perf_metric":152,"perf_label":186,"perf_value":187,"workload":10,"cost_metric":188},"H200 SXM",11,1128,3.5,20440,[180],"lambda",4800,989,0.4,0.2197,"balanced","21333.3 tok\u002Fs",21333.3,{"effective_value":189,"effective_unit":152,"capacity_factor":157},14933.3,{"gpu_model":191,"gpu_model_id":192,"gpu_count":40,"total_vram_gb":161,"hour_price":193,"monthly_cost":194,"providers":195,"bandwidth_gbps":197,"fp16_tflops":198,"interconnect_score":149,"parallel_efficiency":40,"score":199,"tier":151,"perf_metric":152,"perf_label":200,"perf_value":201,"workload":10,"cost_metric":202},"RTX 3090",3,0.0678,49.49,[145,196],"tensordock",936,71,0.655,"1300 tok\u002Fs",1300,{"effective_value":203,"effective_unit":152,"capacity_factor":157},910,[],false,"value",[208,212,220,227,235],{"name":191,"vram_gb":161,"cheapest_hour":193,"cost":209,"provider":145,"estimated_tps":201},{"hourly":193,"daily":210,"monthly":194,"yearly":211},1.63,593.93,{"name":213,"vram_gb":161,"cheapest_hour":214,"cost":215,"provider":145,"estimated_tps":219},"RTX 4090",0.1344,{"hourly":214,"daily":216,"monthly":217,"yearly":218},3.23,98.11,1177.34,1400,{"name":221,"vram_gb":161,"cheapest_hour":222,"cost":223,"provider":145,"estimated_tps":219},"RTX 4090D",0.1602,{"hourly":222,"daily":224,"monthly":225,"yearly":226},3.84,116.95,1403.35,{"name":228,"vram_gb":161,"cheapest_hour":229,"cost":230,"provider":145,"estimated_tps":234},"A10",0.2015,{"hourly":229,"daily":231,"monthly":232,"yearly":233},4.84,147.1,1765.14,833.3,{"name":236,"vram_gb":161,"cheapest_hour":237,"cost":238,"provider":145,"estimated_tps":242},"L4",0.2022,{"hourly":237,"daily":239,"monthly":240,"yearly":241},4.85,147.61,1771.27,416.7,{"posts_count":41,"benchmarks_count":41}]