[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"model-bge-large-en-v1-5":3},{"id":4,"slug":5,"name":6,"hf_id":6,"organization":7,"architecture":8,"architecture_config":9,"parameters":13,"parameters_label":14,"context_length":15,"license":16,"downloads":17,"likes":18,"status":19,"description":16,"huggingface_url":20,"created_at":21,"seo":22,"category":25,"family":28,"score":31,"variants":32,"quantizations":33,"requirements":60,"gpu_recommendations":123,"fitting_gpus":164,"community":195},450,"bge-large-en-v1-5","BAAI\u002Fbge-large-en-v1.5","BAAI","bert",{"layers":10,"head_dim":11,"hidden_size":12},24,64,1024,333244416,"333.2M",512,null,12413978,713,"published","https:\u002F\u002Fhuggingface.co\u002FBAAI\u002Fbge-large-en-v1.5","2026-08-13T18:02:40+00:00",{"title":6,"description":23,"image":16,"robots":24,"canonical_url":16},"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":26,"slug":27},"Embedding \u002F RAG","embedding",{"name":29,"slug":30},"BGE","bge",0.0288,[],[34,39,44,49,53,57],{"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":46,"size_factor":47,"quality_loss":48},"FP8",8,0.5,0.01,{"format":50,"bits":36,"size_factor":51,"quality_loss":52},"GGUF",0.25,0.05,{"format":54,"bits":36,"size_factor":55,"quality_loss":56},"GPTQ",0.275,0.04,{"format":58,"bits":46,"size_factor":47,"quality_loss":59},"INT8",0.02,{"FP16":61,"FP8":82,"INT8":93,"AWQ":97,"GPTQ":108,"GGUF":112},{"minimum":62,"production":71,"recommended":76},{"scenario":63,"context_length":64,"batch_size":42,"weight_gb":65,"kv_cache_gb":43,"vram_gb":66,"ram_gb":67,"disk_gb":68,"requirement_source":69,"confidence":70},"minimum",4096,0.62,2.1,32,39.97,"inferred",0.95,{"scenario":72,"context_length":73,"batch_size":36,"weight_gb":65,"kv_cache_gb":43,"vram_gb":74,"ram_gb":67,"disk_gb":75,"requirement_source":69,"confidence":70},"production",16384,2.65,104.97,{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":65,"kv_cache_gb":43,"vram_gb":80,"ram_gb":67,"disk_gb":81,"requirement_source":69,"confidence":70},"recommended",8192,2,2.31,65.97,{"minimum":83,"production":87,"recommended":90},{"scenario":63,"context_length":64,"batch_size":42,"weight_gb":84,"kv_cache_gb":43,"vram_gb":85,"ram_gb":67,"disk_gb":86,"requirement_source":69,"confidence":70},0.31,1.65,39.48,{"scenario":72,"context_length":73,"batch_size":36,"weight_gb":84,"kv_cache_gb":43,"vram_gb":88,"ram_gb":67,"disk_gb":89,"requirement_source":69,"confidence":70},2.16,104.48,{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":84,"kv_cache_gb":43,"vram_gb":91,"ram_gb":67,"disk_gb":92,"requirement_source":69,"confidence":70},1.84,65.48,{"minimum":94,"production":95,"recommended":96},{"scenario":63,"context_length":64,"batch_size":42,"weight_gb":84,"kv_cache_gb":43,"vram_gb":85,"ram_gb":67,"disk_gb":86,"requirement_source":69,"confidence":70},{"scenario":72,"context_length":73,"batch_size":36,"weight_gb":84,"kv_cache_gb":43,"vram_gb":88,"ram_gb":67,"disk_gb":89,"requirement_source":69,"confidence":70},{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":84,"kv_cache_gb":43,"vram_gb":91,"ram_gb":67,"disk_gb":92,"requirement_source":69,"confidence":70},{"minimum":98,"production":102,"recommended":105},{"scenario":63,"context_length":64,"batch_size":42,"weight_gb":99,"kv_cache_gb":43,"vram_gb":100,"ram_gb":67,"disk_gb":101,"requirement_source":69,"confidence":70},0.16,1.44,39.25,{"scenario":72,"context_length":73,"batch_size":36,"weight_gb":99,"kv_cache_gb":43,"vram_gb":103,"ram_gb":67,"disk_gb":104,"requirement_source":69,"confidence":70},1.93,104.25,{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":99,"kv_cache_gb":43,"vram_gb":106,"ram_gb":67,"disk_gb":107,"requirement_source":69,"confidence":70},1.62,65.25,{"minimum":109,"production":110,"recommended":111},{"scenario":63,"context_length":64,"batch_size":42,"weight_gb":99,"kv_cache_gb":43,"vram_gb":100,"ram_gb":67,"disk_gb":101,"requirement_source":69,"confidence":70},{"scenario":72,"context_length":73,"batch_size":36,"weight_gb":99,"kv_cache_gb":43,"vram_gb":103,"ram_gb":67,"disk_gb":104,"requirement_source":69,"confidence":70},{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":99,"kv_cache_gb":43,"vram_gb":106,"ram_gb":67,"disk_gb":107,"requirement_source":69,"confidence":70},{"minimum":113,"production":117,"recommended":120},{"scenario":63,"context_length":64,"batch_size":42,"weight_gb":114,"kv_cache_gb":43,"vram_gb":115,"ram_gb":67,"disk_gb":116,"requirement_source":69,"confidence":70},0.17,1.45,39.26,{"scenario":72,"context_length":73,"batch_size":36,"weight_gb":114,"kv_cache_gb":43,"vram_gb":118,"ram_gb":67,"disk_gb":119,"requirement_source":69,"confidence":70},1.94,104.26,{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":114,"kv_cache_gb":43,"vram_gb":121,"ram_gb":67,"disk_gb":122,"requirement_source":69,"confidence":70},1.63,65.26,{"recommended":124,"cheapest":140,"performance":143,"min_complexity":153,"cluster_options":161,"no_match":162,"required_vram_gb":106,"candidate_count":36,"usage":163},{"gpu_model":125,"gpu_model_id":10,"gpu_count":42,"total_vram_gb":46,"hour_price":126,"monthly_cost":127,"providers":128,"bandwidth_gbps":130,"fp16_tflops":10,"interconnect_score":131,"parallel_efficiency":42,"score":132,"tier":133,"perf_metric":134,"perf_label":135,"perf_value":136,"workload":16,"cost_metric":137},"RTX 5060 Ti",0.0554,40.44,[129],"vast",448,40,0.5464,"cheapest","vectors\u002Fs","5600 vec\u002Fs",5600,{"effective_value":138,"effective_unit":134,"capacity_factor":139},4480,0.8,{"gpu_model":125,"gpu_model_id":10,"gpu_count":42,"total_vram_gb":46,"hour_price":126,"monthly_cost":127,"providers":141,"bandwidth_gbps":130,"fp16_tflops":10,"interconnect_score":131,"parallel_efficiency":42,"score":132,"tier":133,"perf_metric":134,"perf_label":135,"perf_value":136,"workload":16,"cost_metric":142},[129],{"effective_value":138,"effective_unit":134,"capacity_factor":139},{"gpu_model":125,"gpu_model_id":10,"gpu_count":46,"total_vram_gb":11,"hour_price":126,"monthly_cost":144,"providers":145,"bandwidth_gbps":130,"fp16_tflops":10,"interconnect_score":131,"parallel_efficiency":146,"score":147,"tier":148,"perf_metric":134,"perf_label":149,"perf_value":150,"workload":16,"cost_metric":151},323.54,[129],0.4,0.3635,"balanced","44800 vec\u002Fs",44800,{"effective_value":152,"effective_unit":134,"capacity_factor":139},35840,{"gpu_model":125,"gpu_model_id":10,"gpu_count":79,"total_vram_gb":41,"hour_price":126,"monthly_cost":154,"providers":155,"bandwidth_gbps":130,"fp16_tflops":10,"interconnect_score":131,"parallel_efficiency":139,"score":156,"tier":133,"perf_metric":134,"perf_label":157,"perf_value":158,"workload":16,"cost_metric":159},80.88,[129],0.4509,"11200 vec\u002Fs",11200,{"effective_value":160,"effective_unit":134,"capacity_factor":139},8960,[],false,"value",[165,172,177,184,187],{"name":166,"vram_gb":41,"cheapest_hour":167,"cost":168,"provider":129,"estimated_tps":43},"Tesla V100",0.0272,{"hourly":167,"daily":169,"monthly":170,"yearly":171},0.65,19.86,238.27,{"name":125,"vram_gb":46,"cheapest_hour":126,"cost":173,"provider":129,"estimated_tps":176},{"hourly":126,"daily":174,"monthly":127,"yearly":175},1.33,485.3,2800,{"name":178,"vram_gb":10,"cheapest_hour":179,"cost":180,"provider":129,"estimated_tps":183},"RTX 3090",0.0678,{"hourly":179,"daily":121,"monthly":181,"yearly":182},49.49,593.93,5850,{"name":185,"vram_gb":41,"cheapest_hour":179,"cost":186,"provider":129,"estimated_tps":43},"RTX 4070S Ti",{"hourly":179,"daily":121,"monthly":181,"yearly":182},{"name":188,"vram_gb":189,"cheapest_hour":190,"cost":191,"provider":129,"estimated_tps":194},"RTX 5070",12,0.0804,{"hourly":190,"daily":103,"monthly":192,"yearly":193},58.69,704.3,4200,{"posts_count":43,"benchmarks_count":43}]