[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"model-multilingual-e5-base":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":122,"fitting_gpus":164,"community":196},481,"multilingual-e5-base","intfloat\u002Fmultilingual-e5-base","intfloat","xlm-roberta",{"layers":10,"head_dim":11,"hidden_size":12},12,64,768,276936192,"276.9M",514,null,6436155,379,"published","https:\u002F\u002Fhuggingface.co\u002Fintfloat\u002Fmultilingual-e5-base","2026-08-13T18:08:13+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},"Yi","yi",0.0143,[],[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":107,"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",4096,0.52,1.95,32,39.8,"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",16384,2.48,104.8,{"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",8192,2,2.15,65.8,{"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},0.26,1.58,39.4,{"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},2.08,104.4,{"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},1.77,65.4,{"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},0.13,1.4,39.21,{"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},1.89,104.21,{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":99,"kv_cache_gb":43,"vram_gb":85,"ram_gb":67,"disk_gb":106,"requirement_source":69,"confidence":70},65.21,{"minimum":108,"production":110,"recommended":111},{"scenario":63,"context_length":64,"batch_size":42,"weight_gb":109,"kv_cache_gb":43,"vram_gb":100,"ram_gb":67,"disk_gb":101,"requirement_source":69,"confidence":70},0.14,{"scenario":72,"context_length":73,"batch_size":36,"weight_gb":109,"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":109,"kv_cache_gb":43,"vram_gb":85,"ram_gb":67,"disk_gb":106,"requirement_source":69,"confidence":70},{"minimum":113,"production":116,"recommended":119},{"scenario":63,"context_length":64,"batch_size":42,"weight_gb":109,"kv_cache_gb":43,"vram_gb":114,"ram_gb":67,"disk_gb":115,"requirement_source":69,"confidence":70},1.41,39.22,{"scenario":72,"context_length":73,"batch_size":36,"weight_gb":109,"kv_cache_gb":43,"vram_gb":117,"ram_gb":67,"disk_gb":118,"requirement_source":69,"confidence":70},1.9,104.22,{"scenario":77,"context_length":78,"batch_size":79,"weight_gb":109,"kv_cache_gb":43,"vram_gb":120,"ram_gb":67,"disk_gb":121,"requirement_source":69,"confidence":70},1.59,65.22,{"recommended":123,"cheapest":140,"performance":143,"min_complexity":153,"cluster_options":161,"no_match":162,"required_vram_gb":85,"candidate_count":36,"usage":163},{"gpu_model":124,"gpu_model_id":125,"gpu_count":42,"total_vram_gb":46,"hour_price":126,"monthly_cost":127,"providers":128,"bandwidth_gbps":130,"fp16_tflops":125,"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",24,0.0554,40.44,[129],"vast",448,40,0.545,"cheapest","vectors\u002Fs","6892 vec\u002Fs",6892,{"effective_value":138,"effective_unit":134,"capacity_factor":139},5513.6,0.8,{"gpu_model":124,"gpu_model_id":125,"gpu_count":42,"total_vram_gb":46,"hour_price":126,"monthly_cost":127,"providers":141,"bandwidth_gbps":130,"fp16_tflops":125,"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":124,"gpu_model_id":125,"gpu_count":46,"total_vram_gb":11,"hour_price":126,"monthly_cost":144,"providers":145,"bandwidth_gbps":130,"fp16_tflops":125,"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","55138 vec\u002Fs",55138,{"effective_value":152,"effective_unit":134,"capacity_factor":139},44110.4,{"gpu_model":124,"gpu_model_id":125,"gpu_count":79,"total_vram_gb":41,"hour_price":126,"monthly_cost":154,"providers":155,"bandwidth_gbps":130,"fp16_tflops":125,"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,"13785 vec\u002Fs",13785,{"effective_value":160,"effective_unit":134,"capacity_factor":139},11028,[],false,"value",[165,172,177,185,188],{"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":124,"vram_gb":46,"cheapest_hour":126,"cost":173,"provider":129,"estimated_tps":176},{"hourly":126,"daily":174,"monthly":127,"yearly":175},1.33,485.3,3446.2,{"name":178,"vram_gb":125,"cheapest_hour":179,"cost":180,"provider":129,"estimated_tps":184},"RTX 3090",0.0678,{"hourly":179,"daily":181,"monthly":182,"yearly":183},1.63,49.49,593.93,7200,{"name":186,"vram_gb":41,"cheapest_hour":179,"cost":187,"provider":129,"estimated_tps":43},"RTX 4070S Ti",{"hourly":179,"daily":181,"monthly":182,"yearly":183},{"name":189,"vram_gb":10,"cheapest_hour":190,"cost":191,"provider":129,"estimated_tps":195},"RTX 5070",0.0804,{"hourly":190,"daily":192,"monthly":193,"yearly":194},1.93,58.69,704.3,5169.2,{"posts_count":43,"benchmarks_count":43}]