[{"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":193,"community":217},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":160,"min_complexity":176,"cluster_options":190,"no_match":191,"required_vram_gb":106,"candidate_count":153,"usage":192},{"gpu_model":134,"gpu_model_id":135,"gpu_count":40,"total_vram_gb":39,"hour_price":136,"monthly_cost":137,"providers":138,"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 5080",23,0.1078,78.69,[139],"vast",960,113,40,0.6549,"cheapest","tok\u002Fs","331 tok\u002Fs",331,{"effective_value":149,"effective_unit":145,"capacity_factor":150},231.7,0.7,{"gpu_model":152,"gpu_model_id":153,"gpu_count":40,"total_vram_gb":39,"hour_price":154,"monthly_cost":155,"providers":156,"bandwidth_gbps":41,"fp16_tflops":41,"interconnect_score":142,"parallel_efficiency":40,"score":157,"tier":144,"perf_metric":145,"perf_label":158,"perf_value":41,"workload":10,"cost_metric":159},"Tesla V100",17,0.0272,19.86,[139],0.6493,"0 tok\u002Fs",{"effective_value":41,"effective_unit":145,"capacity_factor":150},{"gpu_model":161,"gpu_model_id":72,"gpu_count":44,"total_vram_gb":162,"hour_price":163,"monthly_cost":164,"providers":165,"bandwidth_gbps":167,"fp16_tflops":168,"interconnect_score":142,"parallel_efficiency":169,"score":170,"tier":171,"perf_metric":145,"perf_label":172,"perf_value":173,"workload":10,"cost_metric":174},"RTX 5090",256,0.3222,1881.65,[139,166],"runpod",1792,209,0.4,0.3618,"balanced","1977.4 tok\u002Fs",1977.4,{"effective_value":175,"effective_unit":145,"capacity_factor":150},1384.2,{"gpu_model":177,"gpu_model_id":122,"gpu_count":40,"total_vram_gb":178,"hour_price":179,"monthly_cost":180,"providers":181,"bandwidth_gbps":183,"fp16_tflops":184,"interconnect_score":142,"parallel_efficiency":40,"score":185,"tier":144,"perf_metric":145,"perf_label":186,"perf_value":187,"workload":10,"cost_metric":188},"RTX 3090",24,0.0678,49.49,[139,182],"tensordock",936,71,0.6021,"322.8 tok\u002Fs",322.8,{"effective_value":189,"effective_unit":145,"capacity_factor":150},226,[],false,"value",[194,198,202,205,213],{"name":152,"vram_gb":39,"cheapest_hour":154,"cost":195,"provider":139,"estimated_tps":41},{"hourly":154,"daily":196,"monthly":155,"yearly":197},0.65,238.27,{"name":177,"vram_gb":178,"cheapest_hour":179,"cost":199,"provider":139,"estimated_tps":187},{"hourly":179,"daily":200,"monthly":180,"yearly":201},1.63,593.93,{"name":203,"vram_gb":39,"cheapest_hour":179,"cost":204,"provider":139,"estimated_tps":41},"RTX 4070S Ti",{"hourly":179,"daily":200,"monthly":180,"yearly":201},{"name":206,"vram_gb":207,"cheapest_hour":208,"cost":209,"provider":139,"estimated_tps":149},"RTX 5070",12,0.0804,{"hourly":208,"daily":210,"monthly":211,"yearly":212},1.93,58.69,704.3,{"name":134,"vram_gb":39,"cheapest_hour":136,"cost":214,"provider":139,"estimated_tps":147},{"hourly":136,"daily":215,"monthly":137,"yearly":216},2.59,944.33,{"posts_count":41,"benchmarks_count":41}]