[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"model-qwen3-0-6b":3},{"id":4,"slug":5,"name":6,"hf_id":6,"organization":7,"architecture":8,"architecture_config":9,"parameters":14,"parameters_label":15,"context_length":16,"license":17,"downloads":18,"likes":19,"status":20,"description":17,"huggingface_url":21,"created_at":22,"seo":23,"category":26,"family":29,"score":31,"variants":32,"quantizations":33,"requirements":59,"gpu_recommendations":129,"fitting_gpus":193,"community":224},438,"qwen3-0-6b","Qwen\u002FQwen3-0.6B","Qwen","qwen3",{"layers":10,"head_dim":11,"kv_heads":12,"hidden_size":13},28,128,8,1024,507904000,"507.9M",40960,null,29660008,1511,"published","https:\u002F\u002Fhuggingface.co\u002FQwen\u002FQwen3-0.6B","2026-08-13T18:01:12+00:00",{"title":6,"description":24,"image":17,"robots":25,"canonical_url":17},"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":27,"slug":28},"LLM","llm",{"name":7,"slug":30},"qwen",0.0632,[],[34,39,44,48,52,56],{"format":35,"bits":36,"size_factor":37,"quality_loss":38},"AWQ",4,0.263,0.03,{"format":40,"bits":41,"size_factor":42,"quality_loss":43},"FP16",16,1,0,{"format":45,"bits":12,"size_factor":46,"quality_loss":47},"FP8",0.5,0.01,{"format":49,"bits":36,"size_factor":50,"quality_loss":51},"GGUF",0.25,0.05,{"format":53,"bits":36,"size_factor":54,"quality_loss":55},"GPTQ",0.275,0.04,{"format":57,"bits":12,"size_factor":46,"quality_loss":58},"INT8",0.02,{"FP16":60,"FP8":84,"INT8":96,"AWQ":100,"GPTQ":111,"GGUF":119},{"minimum":61,"production":70,"recommended":77},{"scenario":62,"context_length":63,"batch_size":42,"weight_gb":64,"kv_cache_gb":65,"vram_gb":66,"ram_gb":67,"disk_gb":68,"requirement_source":69,"confidence":64},"minimum",4096,0.95,0.44,4.36,32,40.48,"inferred",{"scenario":71,"context_length":72,"batch_size":36,"weight_gb":64,"kv_cache_gb":73,"vram_gb":74,"ram_gb":75,"disk_gb":76,"requirement_source":69,"confidence":64},"production",16384,56,79.31,118.96,105.48,{"scenario":78,"context_length":79,"batch_size":80,"weight_gb":64,"kv_cache_gb":81,"vram_gb":82,"ram_gb":67,"disk_gb":83,"requirement_source":69,"confidence":64},"recommended",8192,2,7,13.61,66.48,{"minimum":85,"production":89,"recommended":93},{"scenario":62,"context_length":63,"batch_size":42,"weight_gb":86,"kv_cache_gb":65,"vram_gb":87,"ram_gb":67,"disk_gb":88,"requirement_source":69,"confidence":64},0.47,3.76,39.74,{"scenario":71,"context_length":72,"batch_size":36,"weight_gb":86,"kv_cache_gb":73,"vram_gb":90,"ram_gb":91,"disk_gb":92,"requirement_source":69,"confidence":64},78.65,117.98,104.74,{"scenario":78,"context_length":79,"batch_size":80,"weight_gb":86,"kv_cache_gb":81,"vram_gb":94,"ram_gb":67,"disk_gb":95,"requirement_source":69,"confidence":64},12.99,65.74,{"minimum":97,"production":98,"recommended":99},{"scenario":62,"context_length":63,"batch_size":42,"weight_gb":86,"kv_cache_gb":65,"vram_gb":87,"ram_gb":67,"disk_gb":88,"requirement_source":69,"confidence":64},{"scenario":71,"context_length":72,"batch_size":36,"weight_gb":86,"kv_cache_gb":73,"vram_gb":90,"ram_gb":91,"disk_gb":92,"requirement_source":69,"confidence":64},{"scenario":78,"context_length":79,"batch_size":80,"weight_gb":86,"kv_cache_gb":81,"vram_gb":94,"ram_gb":67,"disk_gb":95,"requirement_source":69,"confidence":64},{"minimum":101,"production":104,"recommended":108},{"scenario":62,"context_length":63,"batch_size":42,"weight_gb":50,"kv_cache_gb":65,"vram_gb":102,"ram_gb":67,"disk_gb":103,"requirement_source":69,"confidence":64},3.47,39.38,{"scenario":71,"context_length":72,"batch_size":36,"weight_gb":50,"kv_cache_gb":73,"vram_gb":105,"ram_gb":106,"disk_gb":107,"requirement_source":69,"confidence":64},78.34,117.51,104.38,{"scenario":78,"context_length":79,"batch_size":80,"weight_gb":50,"kv_cache_gb":81,"vram_gb":109,"ram_gb":67,"disk_gb":110,"requirement_source":69,"confidence":64},12.69,65.38,{"minimum":112,"production":115,"recommended":117},{"scenario":62,"context_length":63,"batch_size":42,"weight_gb":50,"kv_cache_gb":65,"vram_gb":113,"ram_gb":67,"disk_gb":114,"requirement_source":69,"confidence":64},3.48,39.39,{"scenario":71,"context_length":72,"batch_size":36,"weight_gb":50,"kv_cache_gb":73,"vram_gb":105,"ram_gb":106,"disk_gb":116,"requirement_source":69,"confidence":64},104.39,{"scenario":78,"context_length":79,"batch_size":80,"weight_gb":50,"kv_cache_gb":81,"vram_gb":109,"ram_gb":67,"disk_gb":118,"requirement_source":69,"confidence":64},65.39,{"minimum":120,"production":122,"recommended":126},{"scenario":62,"context_length":63,"batch_size":42,"weight_gb":50,"kv_cache_gb":65,"vram_gb":113,"ram_gb":67,"disk_gb":121,"requirement_source":69,"confidence":64},39.4,{"scenario":71,"context_length":72,"batch_size":36,"weight_gb":50,"kv_cache_gb":73,"vram_gb":123,"ram_gb":124,"disk_gb":125,"requirement_source":69,"confidence":64},78.35,117.53,104.4,{"scenario":78,"context_length":79,"batch_size":80,"weight_gb":50,"kv_cache_gb":81,"vram_gb":127,"ram_gb":67,"disk_gb":128,"requirement_source":69,"confidence":64},12.7,65.4,{"recommended":130,"cheapest":148,"performance":157,"min_complexity":174,"cluster_options":189,"no_match":190,"required_vram_gb":109,"candidate_count":191,"usage":192},{"gpu_model":131,"gpu_model_id":80,"gpu_count":42,"total_vram_gb":67,"hour_price":132,"monthly_cost":133,"providers":134,"bandwidth_gbps":137,"fp16_tflops":138,"interconnect_score":139,"parallel_efficiency":42,"score":140,"tier":141,"perf_metric":142,"perf_label":143,"perf_value":144,"workload":17,"cost_metric":145},"RTX 5090",0.3337,243.6,[135,136],"vast","runpod",1792,209,40,0.6035,"cheapest","tok\u002Fs","7168 tok\u002Fs",7168,{"effective_value":146,"effective_unit":142,"capacity_factor":147},5017.6,0.7,{"gpu_model":149,"gpu_model_id":150,"gpu_count":42,"total_vram_gb":41,"hour_price":151,"monthly_cost":152,"providers":153,"bandwidth_gbps":43,"fp16_tflops":43,"interconnect_score":139,"parallel_efficiency":42,"score":154,"tier":141,"perf_metric":142,"perf_label":155,"perf_value":43,"workload":17,"cost_metric":156},"Tesla V100",17,0.0289,21.1,[135],0.6781,"0 tok\u002Fs",{"effective_value":43,"effective_unit":142,"capacity_factor":147},{"gpu_model":158,"gpu_model_id":159,"gpu_count":12,"total_vram_gb":160,"hour_price":161,"monthly_cost":162,"providers":163,"bandwidth_gbps":165,"fp16_tflops":166,"interconnect_score":139,"parallel_efficiency":167,"score":168,"tier":169,"perf_metric":142,"perf_label":170,"perf_value":171,"workload":17,"cost_metric":172},"H200 SXM",11,1128,3.5,20440,[164],"lambda",4800,989,0.4,0.2197,"balanced","61440 tok\u002Fs",61440,{"effective_value":173,"effective_unit":142,"capacity_factor":147},43008,{"gpu_model":175,"gpu_model_id":176,"gpu_count":42,"total_vram_gb":177,"hour_price":178,"monthly_cost":179,"providers":180,"bandwidth_gbps":182,"fp16_tflops":183,"interconnect_score":139,"parallel_efficiency":42,"score":184,"tier":141,"perf_metric":142,"perf_label":185,"perf_value":186,"workload":17,"cost_metric":187},"RTX 3090",3,24,0.0622,45.41,[135,181],"tensordock",936,71,0.6367,"3744 tok\u002Fs",3744,{"effective_value":188,"effective_unit":142,"capacity_factor":147},2620.8,[],false,26,"value",[194,198,202,209,216],{"name":149,"vram_gb":41,"cheapest_hour":151,"cost":195,"provider":135,"estimated_tps":43},{"hourly":151,"daily":196,"monthly":152,"yearly":197},0.69,253.16,{"name":175,"vram_gb":177,"cheapest_hour":178,"cost":199,"provider":135,"estimated_tps":186},{"hourly":178,"daily":200,"monthly":179,"yearly":201},1.49,544.87,{"name":203,"vram_gb":41,"cheapest_hour":204,"cost":205,"provider":135,"estimated_tps":43},"RTX 4070S Ti",0.0678,{"hourly":204,"daily":206,"monthly":207,"yearly":208},1.63,49.49,593.93,{"name":210,"vram_gb":41,"cheapest_hour":211,"cost":212,"provider":135,"estimated_tps":43},"RTX 4080S",0.0685,{"hourly":211,"daily":213,"monthly":214,"yearly":215},1.64,50.01,600.06,{"name":217,"vram_gb":41,"cheapest_hour":218,"cost":219,"provider":135,"estimated_tps":223},"RTX 5070 Ti",0.0849,{"hourly":218,"daily":220,"monthly":221,"yearly":222},2.04,61.98,743.72,3584,{"posts_count":43,"benchmarks_count":43}]