[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"model-yi-vl-6b":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":132,"fitting_gpus":192,"community":219},411,"yi-vl-6b","Yi-VL 6B","01-ai\u002FYi-VL-6B","01-ai","transformer",null,6000000000,"6B",4096,"MIT",249,124,"published","https:\u002F\u002Fhuggingface.co\u002F01-ai\u002FYi-VL-6B","2026-08-12T03:11:27+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},"OCR \u002F Vision","ocr-vision",{"name":27,"slug":28},"Yi","yi",0.0014,[],[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":80,"INT8":92,"AWQ":96,"GPTQ":108,"GGUF":120},{"minimum":60,"production":69,"recommended":76},{"scenario":61,"context_length":62,"batch_size":40,"weight_gb":63,"kv_cache_gb":49,"vram_gb":64,"ram_gb":65,"disk_gb":66,"requirement_source":67,"confidence":68},"minimum",2048,11.18,17.71,32,56.43,"estimated",0.95,{"scenario":70,"context_length":71,"batch_size":72,"weight_gb":63,"kv_cache_gb":44,"vram_gb":73,"ram_gb":74,"disk_gb":75,"requirement_source":67,"confidence":68},"production",8192,2,37.02,55.53,121.43,{"scenario":77,"context_length":13,"batch_size":40,"weight_gb":63,"kv_cache_gb":40,"vram_gb":78,"ram_gb":65,"disk_gb":79,"requirement_source":67,"confidence":68},"recommended",20.53,82.43,{"minimum":81,"production":85,"recommended":89},{"scenario":61,"context_length":62,"batch_size":40,"weight_gb":82,"kv_cache_gb":49,"vram_gb":83,"ram_gb":65,"disk_gb":84,"requirement_source":67,"confidence":68},5.59,10.64,47.72,{"scenario":70,"context_length":71,"batch_size":72,"weight_gb":82,"kv_cache_gb":44,"vram_gb":86,"ram_gb":87,"disk_gb":88,"requirement_source":67,"confidence":68},29.31,43.97,112.72,{"scenario":77,"context_length":13,"batch_size":40,"weight_gb":82,"kv_cache_gb":40,"vram_gb":90,"ram_gb":65,"disk_gb":91,"requirement_source":67,"confidence":68},13.14,73.72,{"minimum":93,"production":94,"recommended":95},{"scenario":61,"context_length":62,"batch_size":40,"weight_gb":82,"kv_cache_gb":49,"vram_gb":83,"ram_gb":65,"disk_gb":84,"requirement_source":67,"confidence":68},{"scenario":70,"context_length":71,"batch_size":72,"weight_gb":82,"kv_cache_gb":44,"vram_gb":86,"ram_gb":87,"disk_gb":88,"requirement_source":67,"confidence":68},{"scenario":77,"context_length":13,"batch_size":40,"weight_gb":82,"kv_cache_gb":40,"vram_gb":90,"ram_gb":65,"disk_gb":91,"requirement_source":67,"confidence":68},{"minimum":97,"production":101,"recommended":105},{"scenario":61,"context_length":62,"batch_size":40,"weight_gb":98,"kv_cache_gb":49,"vram_gb":99,"ram_gb":65,"disk_gb":100,"requirement_source":67,"confidence":68},2.9,7.24,43.52,{"scenario":70,"context_length":71,"batch_size":72,"weight_gb":98,"kv_cache_gb":44,"vram_gb":102,"ram_gb":103,"disk_gb":104,"requirement_source":67,"confidence":68},25.6,38.4,108.52,{"scenario":77,"context_length":13,"batch_size":40,"weight_gb":98,"kv_cache_gb":40,"vram_gb":106,"ram_gb":65,"disk_gb":107,"requirement_source":67,"confidence":68},9.58,69.52,{"minimum":109,"production":113,"recommended":117},{"scenario":61,"context_length":62,"batch_size":40,"weight_gb":110,"kv_cache_gb":49,"vram_gb":111,"ram_gb":65,"disk_gb":112,"requirement_source":67,"confidence":68},2.93,7.29,43.58,{"scenario":70,"context_length":71,"batch_size":72,"weight_gb":110,"kv_cache_gb":44,"vram_gb":114,"ram_gb":115,"disk_gb":116,"requirement_source":67,"confidence":68},25.65,38.47,108.58,{"scenario":77,"context_length":13,"batch_size":40,"weight_gb":110,"kv_cache_gb":40,"vram_gb":118,"ram_gb":65,"disk_gb":119,"requirement_source":67,"confidence":68},9.63,69.58,{"minimum":121,"production":125,"recommended":129},{"scenario":61,"context_length":62,"batch_size":40,"weight_gb":122,"kv_cache_gb":49,"vram_gb":123,"ram_gb":65,"disk_gb":124,"requirement_source":67,"confidence":68},3,7.37,43.69,{"scenario":70,"context_length":71,"batch_size":72,"weight_gb":122,"kv_cache_gb":44,"vram_gb":126,"ram_gb":127,"disk_gb":128,"requirement_source":67,"confidence":68},25.74,38.62,108.69,{"scenario":77,"context_length":13,"batch_size":40,"weight_gb":122,"kv_cache_gb":40,"vram_gb":130,"ram_gb":65,"disk_gb":131,"requirement_source":67,"confidence":68},9.72,69.69,{"recommended":133,"cheapest":151,"performance":164,"min_complexity":175,"cluster_options":189,"no_match":190,"required_vram_gb":106,"candidate_count":39,"usage":191},{"gpu_model":134,"gpu_model_id":72,"gpu_count":40,"total_vram_gb":65,"hour_price":135,"monthly_cost":136,"providers":137,"bandwidth_gbps":140,"fp16_tflops":141,"interconnect_score":142,"parallel_efficiency":40,"score":143,"tier":144,"perf_metric":145,"perf_label":146,"perf_value":147,"workload":10,"cost_metric":148},"RTX 5090",0.337,246.01,[138,139],"vast","runpod",1792,209,40,0.5775,"cheapest","tok\u002Fs","617.9 tok\u002Fs",617.9,{"effective_value":149,"effective_unit":145,"capacity_factor":150},432.5,0.7,{"gpu_model":152,"gpu_model_id":153,"gpu_count":40,"total_vram_gb":154,"hour_price":155,"monthly_cost":123,"providers":156,"bandwidth_gbps":157,"fp16_tflops":158,"interconnect_score":142,"parallel_efficiency":40,"score":159,"tier":144,"perf_metric":145,"perf_label":160,"perf_value":161,"workload":10,"cost_metric":162},"RTX 5070",28,12,0.0101,[138],672,61,0.6818,"231.7 tok\u002Fs",231.7,{"effective_value":163,"effective_unit":145,"capacity_factor":150},162.2,{"gpu_model":134,"gpu_model_id":72,"gpu_count":44,"total_vram_gb":165,"hour_price":135,"monthly_cost":166,"providers":167,"bandwidth_gbps":140,"fp16_tflops":141,"interconnect_score":142,"parallel_efficiency":168,"score":169,"tier":170,"perf_metric":145,"perf_label":171,"perf_value":172,"workload":10,"cost_metric":173},256,1968.08,[138,139],0.4,0.36,"balanced","1977.4 tok\u002Fs",1977.4,{"effective_value":174,"effective_unit":145,"capacity_factor":150},1384.2,{"gpu_model":176,"gpu_model_id":122,"gpu_count":40,"total_vram_gb":177,"hour_price":178,"monthly_cost":179,"providers":180,"bandwidth_gbps":182,"fp16_tflops":183,"interconnect_score":142,"parallel_efficiency":40,"score":184,"tier":144,"perf_metric":145,"perf_label":185,"perf_value":186,"workload":10,"cost_metric":187},"RTX 3090",24,0.0678,49.49,[138,181],"tensordock",936,71,0.6021,"322.8 tok\u002Fs",322.8,{"effective_value":188,"effective_unit":145,"capacity_factor":150},226,[],false,"value",[193,197,204,208,211],{"name":152,"vram_gb":154,"cheapest_hour":155,"cost":194,"provider":138,"estimated_tps":161},{"hourly":155,"daily":195,"monthly":123,"yearly":196},0.24,88.48,{"name":198,"vram_gb":39,"cheapest_hour":199,"cost":200,"provider":138,"estimated_tps":41},"Tesla V100",0.0272,{"hourly":199,"daily":201,"monthly":202,"yearly":203},0.65,19.86,238.27,{"name":176,"vram_gb":177,"cheapest_hour":178,"cost":205,"provider":138,"estimated_tps":186},{"hourly":178,"daily":206,"monthly":179,"yearly":207},1.63,593.93,{"name":209,"vram_gb":39,"cheapest_hour":178,"cost":210,"provider":138,"estimated_tps":41},"RTX 4070S Ti",{"hourly":178,"daily":206,"monthly":179,"yearly":207},{"name":212,"vram_gb":39,"cheapest_hour":213,"cost":214,"provider":138,"estimated_tps":218},"RTX 5070 Ti",0.0849,{"hourly":213,"daily":215,"monthly":216,"yearly":217},2.04,61.98,743.72,309,{"posts_count":41,"benchmarks_count":41}]