[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"model-deepseek-v3-2":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":32,"variants":33,"quantizations":34,"requirements":61,"gpu_recommendations":146,"fitting_gpus":449,"community":450},116,"deepseek-v3-2","deepseek-ai\u002FDeepSeek-V3.2","deepseek-ai","deepseek_v32",{"layers":10,"head_dim":11,"kv_heads":12,"hidden_size":13},61,56,128,7168,38537003008,"38.5B",163840,null,4668170,1508,"published","https:\u002F\u002Fhuggingface.co\u002Fdeepseek-ai\u002FDeepSeek-V3.2","2026-08-12T03:11:12+00:00",{"title":6,"description":24,"image":17,"robots":25,"canonical_url":17},"AI model deployment & compute intelligence — compare AI models, GPU requirements, benchmarks, cloud GPU rental pricing and deployment cost analysis for DeepSeek, Qwen, Llama, Wan and more.","index, follow",{"name":27,"slug":28},"LLM","llm",{"name":30,"slug":31},"DeepSeek","deepseek",0.0223,[],[35,40,45,50,54,58],{"format":36,"bits":37,"size_factor":38,"quality_loss":39},"AWQ",4,0.263,0.03,{"format":41,"bits":42,"size_factor":43,"quality_loss":44},"FP16",16,1,0,{"format":46,"bits":47,"size_factor":48,"quality_loss":49},"FP8",8,0.5,0.01,{"format":51,"bits":37,"size_factor":52,"quality_loss":53},"GGUF",0.25,0.05,{"format":55,"bits":37,"size_factor":56,"quality_loss":57},"GPTQ",0.275,0.04,{"format":59,"bits":47,"size_factor":48,"quality_loss":60},"INT8",0.02,{"FP16":62,"FP8":86,"INT8":100,"AWQ":104,"GPTQ":118,"GGUF":132},{"minimum":63,"production":72,"recommended":79},{"scenario":64,"context_length":65,"batch_size":43,"weight_gb":66,"kv_cache_gb":48,"vram_gb":67,"ram_gb":68,"disk_gb":69,"requirement_source":70,"confidence":71},"minimum",4096,1249.83,1585.99,2378.99,1988.74,"estimated",0.95,{"scenario":73,"context_length":74,"batch_size":37,"weight_gb":66,"kv_cache_gb":75,"vram_gb":76,"ram_gb":77,"disk_gb":78,"requirement_source":70,"confidence":71},"production",16384,64,1842.37,2763.56,2053.74,{"scenario":80,"context_length":81,"batch_size":82,"weight_gb":66,"kv_cache_gb":47,"vram_gb":83,"ram_gb":84,"disk_gb":85,"requirement_source":70,"confidence":71},"recommended",8192,2,1673.61,2510.41,2014.74,{"minimum":87,"production":92,"recommended":96},{"scenario":64,"context_length":65,"batch_size":43,"weight_gb":88,"kv_cache_gb":48,"vram_gb":89,"ram_gb":90,"disk_gb":91,"requirement_source":70,"confidence":71},624.92,795.47,1193.21,1013.87,{"scenario":73,"context_length":74,"batch_size":37,"weight_gb":88,"kv_cache_gb":75,"vram_gb":93,"ram_gb":94,"disk_gb":95,"requirement_source":70,"confidence":71},979.99,1469.98,1078.87,{"scenario":80,"context_length":81,"batch_size":82,"weight_gb":88,"kv_cache_gb":47,"vram_gb":97,"ram_gb":98,"disk_gb":99,"requirement_source":70,"confidence":71},847.15,1270.73,1039.87,{"minimum":101,"production":102,"recommended":103},{"scenario":64,"context_length":65,"batch_size":43,"weight_gb":88,"kv_cache_gb":48,"vram_gb":89,"ram_gb":90,"disk_gb":91,"requirement_source":70,"confidence":71},{"scenario":73,"context_length":74,"batch_size":37,"weight_gb":88,"kv_cache_gb":75,"vram_gb":93,"ram_gb":94,"disk_gb":95,"requirement_source":70,"confidence":71},{"scenario":80,"context_length":81,"batch_size":82,"weight_gb":88,"kv_cache_gb":47,"vram_gb":97,"ram_gb":98,"disk_gb":99,"requirement_source":70,"confidence":71},{"minimum":105,"production":110,"recommended":114},{"scenario":64,"context_length":65,"batch_size":43,"weight_gb":106,"kv_cache_gb":48,"vram_gb":107,"ram_gb":108,"disk_gb":109,"requirement_source":70,"confidence":71},324.18,415.03,622.55,544.71,{"scenario":73,"context_length":74,"batch_size":37,"weight_gb":106,"kv_cache_gb":75,"vram_gb":111,"ram_gb":112,"disk_gb":113,"requirement_source":70,"confidence":71},564.96,847.44,609.71,{"scenario":80,"context_length":81,"batch_size":82,"weight_gb":106,"kv_cache_gb":47,"vram_gb":115,"ram_gb":116,"disk_gb":117,"requirement_source":70,"confidence":71},449.42,674.13,570.71,{"minimum":119,"production":124,"recommended":128},{"scenario":64,"context_length":65,"batch_size":43,"weight_gb":120,"kv_cache_gb":48,"vram_gb":121,"ram_gb":122,"disk_gb":123,"requirement_source":70,"confidence":71},328.08,419.97,629.96,550.81,{"scenario":73,"context_length":74,"batch_size":37,"weight_gb":120,"kv_cache_gb":75,"vram_gb":125,"ram_gb":126,"disk_gb":127,"requirement_source":70,"confidence":71},570.35,855.53,615.81,{"scenario":80,"context_length":81,"batch_size":82,"weight_gb":120,"kv_cache_gb":47,"vram_gb":129,"ram_gb":130,"disk_gb":131,"requirement_source":70,"confidence":71},454.59,681.88,576.81,{"minimum":133,"production":138,"recommended":142},{"scenario":64,"context_length":65,"batch_size":43,"weight_gb":134,"kv_cache_gb":48,"vram_gb":135,"ram_gb":136,"disk_gb":137,"requirement_source":70,"confidence":71},335.89,429.85,644.78,562.99,{"scenario":73,"context_length":74,"batch_size":37,"weight_gb":134,"kv_cache_gb":75,"vram_gb":139,"ram_gb":140,"disk_gb":141,"requirement_source":70,"confidence":71},581.13,871.7,627.99,{"scenario":80,"context_length":81,"batch_size":82,"weight_gb":134,"kv_cache_gb":47,"vram_gb":143,"ram_gb":144,"disk_gb":145,"requirement_source":70,"confidence":71},464.92,697.38,588.99,{"recommended":147,"cheapest":167,"performance":178,"min_complexity":188,"cluster_options":196,"no_match":447,"required_vram_gb":115,"candidate_count":199,"usage":448},{"gpu_model":148,"gpu_model_id":149,"gpu_count":47,"total_vram_gb":150,"hour_price":151,"monthly_cost":152,"providers":153,"bandwidth_gbps":155,"fp16_tflops":156,"interconnect_score":157,"parallel_efficiency":158,"score":159,"tier":160,"perf_metric":161,"perf_label":162,"perf_value":163,"workload":17,"cost_metric":164},"H200 SXM",11,1128,3.5,20440,[154],"lambda",4800,989,40,0.4,0.326,"balanced","tok\u002Fs","47.4 tok\u002Fs",47.4,{"effective_value":165,"effective_unit":161,"capacity_factor":166},33.2,0.7,{"gpu_model":168,"gpu_model_id":169,"gpu_count":47,"total_vram_gb":170,"hour_price":171,"monthly_cost":172,"providers":173,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"score":175,"tier":160,"perf_metric":161,"perf_label":176,"perf_value":44,"workload":17,"cost_metric":177},"A100 SXM4",39,640,0.3781,2208.1,[174],"vast",0.5013,"0 tok\u002Fs",{"effective_value":44,"effective_unit":161,"capacity_factor":166},{"gpu_model":148,"gpu_model_id":149,"gpu_count":37,"total_vram_gb":179,"hour_price":151,"monthly_cost":180,"providers":181,"bandwidth_gbps":155,"fp16_tflops":156,"interconnect_score":157,"parallel_efficiency":182,"score":183,"tier":160,"perf_metric":161,"perf_label":184,"perf_value":185,"workload":17,"cost_metric":186},564,10220,[154],0.6,0.4569,"35.5 tok\u002Fs",35.5,{"effective_value":187,"effective_unit":161,"capacity_factor":166},24.8,{"gpu_model":189,"gpu_model_id":190,"gpu_count":37,"total_vram_gb":179,"hour_price":191,"monthly_cost":192,"providers":193,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":182,"score":194,"tier":160,"perf_metric":161,"perf_label":176,"perf_value":44,"workload":17,"cost_metric":195},"H200",38,2.5022,7306.42,[174],0.4353,{"effective_value":44,"effective_unit":161,"capacity_factor":166},[197,209,220,227,233,240,250,257,268,277,286,294,301,309,318,325,334,343,350,357,364,371,378,385,392,399,410,418,429,439],{"gpu_model":198,"gpu_model_id":199,"gpu_count":47,"total_vram_gb":200,"hour_price":201,"monthly_cost":202,"providers":203,"bandwidth_gbps":204,"fp16_tflops":205,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":207,"score":208,"tier":160},"RTX A6000",6,384,0.3622,2115.25,[174,154],768,77,true,65.42,0.4802,{"gpu_model":210,"gpu_model_id":211,"gpu_count":47,"total_vram_gb":200,"hour_price":212,"monthly_cost":213,"providers":214,"bandwidth_gbps":217,"fp16_tflops":218,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":207,"score":219,"tier":160},"L40S",12,0.4415,2578.36,[174,215,216],"tensordock","runpod",864,362,0.4727,{"gpu_model":221,"gpu_model_id":222,"gpu_count":47,"total_vram_gb":200,"hour_price":223,"monthly_cost":224,"providers":225,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":207,"score":226,"tier":160},"RTX PRO 6000 WS",19,1.0037,5861.61,[174],0.3845,{"gpu_model":228,"gpu_model_id":157,"gpu_count":47,"total_vram_gb":200,"hour_price":229,"monthly_cost":230,"providers":231,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":207,"score":232,"tier":160},"RTX PRO 6000 Max-Q",1.1215,6549.56,[174],0.3702,{"gpu_model":234,"gpu_model_id":235,"gpu_count":47,"total_vram_gb":200,"hour_price":236,"monthly_cost":237,"providers":238,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":207,"score":239,"tier":160},"RTX PRO 6000 S",20,1.3341,7791.14,[174],0.3444,{"gpu_model":241,"gpu_model_id":82,"gpu_count":47,"total_vram_gb":242,"hour_price":243,"monthly_cost":244,"providers":245,"bandwidth_gbps":246,"fp16_tflops":247,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":248,"score":249,"tier":160},"RTX 5090",256,0.4006,2339.5,[174,216],1792,209,193.42,0.4989,{"gpu_model":251,"gpu_model_id":252,"gpu_count":47,"total_vram_gb":242,"hour_price":253,"monthly_cost":254,"providers":255,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":248,"score":256,"tier":160},"RTX 5090D",42,0.6678,3899.95,[174],0.4254,{"gpu_model":258,"gpu_model_id":259,"gpu_count":47,"total_vram_gb":260,"hour_price":261,"monthly_cost":262,"providers":263,"bandwidth_gbps":264,"fp16_tflops":265,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":266,"score":267,"tier":160},"L4",13,192,0.1044,609.7,[174],300,121,257.42,0.5008,{"gpu_model":269,"gpu_model_id":270,"gpu_count":47,"total_vram_gb":260,"hour_price":271,"monthly_cost":272,"providers":273,"bandwidth_gbps":274,"fp16_tflops":275,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":266,"score":276,"tier":160},"RTX 3090",3,0.1091,637.14,[174,215],936,71,0.5148,{"gpu_model":278,"gpu_model_id":279,"gpu_count":47,"total_vram_gb":260,"hour_price":280,"monthly_cost":281,"providers":282,"bandwidth_gbps":283,"fp16_tflops":284,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":266,"score":285,"tier":160},"A10",7,0.2413,1409.19,[174,216],600,125,0.491,{"gpu_model":287,"gpu_model_id":43,"gpu_count":47,"total_vram_gb":260,"hour_price":288,"monthly_cost":289,"providers":290,"bandwidth_gbps":291,"fp16_tflops":292,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":266,"score":293,"tier":160},"RTX 4090",0.26,1518.4,[215,216,154,174],1008,165,0.4981,{"gpu_model":295,"gpu_model_id":296,"gpu_count":47,"total_vram_gb":260,"hour_price":297,"monthly_cost":298,"providers":299,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":266,"score":300,"tier":160},"RTX PRO 5000",26,0.3756,2193.5,[174],0.461,{"gpu_model":302,"gpu_model_id":303,"gpu_count":47,"total_vram_gb":260,"hour_price":304,"monthly_cost":305,"providers":306,"bandwidth_gbps":291,"fp16_tflops":307,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":266,"score":308,"tier":160},"RTX 4090D",5,0.9215,5381.56,[174],82.6,0.4176,{"gpu_model":310,"gpu_model_id":311,"gpu_count":47,"total_vram_gb":12,"hour_price":312,"monthly_cost":313,"providers":314,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":315,"score":316,"tier":317},"Tesla V100",17,0.0489,285.58,[174],321.42,0.5007,"cheapest",{"gpu_model":319,"gpu_model_id":320,"gpu_count":47,"total_vram_gb":12,"hour_price":321,"monthly_cost":322,"providers":323,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":315,"score":324,"tier":160},"RTX 4070S Ti",22,0.1339,781.98,[174],0.4904,{"gpu_model":326,"gpu_model_id":327,"gpu_count":47,"total_vram_gb":12,"hour_price":328,"monthly_cost":329,"providers":330,"bandwidth_gbps":331,"fp16_tflops":332,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":315,"score":333,"tier":160},"RTX 5070 Ti",18,0.1622,947.25,[174],896,88,0.5074,{"gpu_model":335,"gpu_model_id":336,"gpu_count":47,"total_vram_gb":12,"hour_price":337,"monthly_cost":338,"providers":339,"bandwidth_gbps":340,"fp16_tflops":341,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":315,"score":342,"tier":160},"RTX 5080",23,0.1889,1103.18,[174],960,113,0.5056,{"gpu_model":344,"gpu_model_id":345,"gpu_count":47,"total_vram_gb":12,"hour_price":346,"monthly_cost":347,"providers":348,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":315,"score":349,"tier":160},"RTX 4080S",21,0.2278,1330.35,[174],0.479,{"gpu_model":351,"gpu_model_id":352,"gpu_count":47,"total_vram_gb":12,"hour_price":353,"monthly_cost":354,"providers":355,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":315,"score":356,"tier":160},"RTX PRO 4000",34,0.2281,1332.1,[174],0.4789,{"gpu_model":358,"gpu_model_id":359,"gpu_count":47,"total_vram_gb":12,"hour_price":360,"monthly_cost":361,"providers":362,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":315,"score":363,"tier":160},"CMP 170HX",37,0.2689,1570.38,[174],0.474,{"gpu_model":365,"gpu_model_id":366,"gpu_count":47,"total_vram_gb":12,"hour_price":367,"monthly_cost":368,"providers":369,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":315,"score":370,"tier":160},"RTX PRO 4500",25,0.3489,2037.58,[174],0.4642,{"gpu_model":372,"gpu_model_id":373,"gpu_count":47,"total_vram_gb":12,"hour_price":374,"monthly_cost":375,"providers":376,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":315,"score":377,"tier":160},"A800 PCIE",44,0.4674,2729.62,[174],0.4498,{"gpu_model":379,"gpu_model_id":380,"gpu_count":47,"total_vram_gb":12,"hour_price":381,"monthly_cost":382,"providers":383,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":315,"score":384,"tier":160},"RTX 6000Ada",41,0.7356,4295.9,[174],0.4172,{"gpu_model":386,"gpu_model_id":387,"gpu_count":47,"total_vram_gb":12,"hour_price":388,"monthly_cost":389,"providers":390,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":315,"score":391,"tier":160},"RTX 6000D",43,1.4689,8578.38,[174],0.328,{"gpu_model":393,"gpu_model_id":394,"gpu_count":47,"total_vram_gb":12,"hour_price":395,"monthly_cost":396,"providers":397,"bandwidth_gbps":44,"fp16_tflops":44,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":315,"score":398,"tier":160},"B300",31,10.626,62055.84,[174],0.2567,{"gpu_model":400,"gpu_model_id":401,"gpu_count":47,"total_vram_gb":402,"hour_price":403,"monthly_cost":404,"providers":405,"bandwidth_gbps":406,"fp16_tflops":407,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":408,"score":409,"tier":160},"RTX 3080 Ti",36,96,0.108,630.72,[174],912,68,353.42,0.5144,{"gpu_model":411,"gpu_model_id":412,"gpu_count":47,"total_vram_gb":402,"hour_price":413,"monthly_cost":414,"providers":415,"bandwidth_gbps":416,"fp16_tflops":10,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":408,"score":417,"tier":160},"RTX 5070",28,0.1211,707.22,[174],672,0.5073,{"gpu_model":419,"gpu_model_id":420,"gpu_count":47,"total_vram_gb":421,"hour_price":422,"monthly_cost":423,"providers":424,"bandwidth_gbps":425,"fp16_tflops":426,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":427,"score":428,"tier":160},"RTX 3080",35,80,0.0874,510.42,[174],760,60,369.42,0.5134,{"gpu_model":430,"gpu_model_id":431,"gpu_count":47,"total_vram_gb":75,"hour_price":432,"monthly_cost":433,"providers":434,"bandwidth_gbps":435,"fp16_tflops":436,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":437,"score":438,"tier":160},"RTX 4060 Ti",29,0.0811,473.62,[174],288,45,385.42,0.5034,{"gpu_model":440,"gpu_model_id":441,"gpu_count":47,"total_vram_gb":75,"hour_price":442,"monthly_cost":443,"providers":444,"bandwidth_gbps":445,"fp16_tflops":441,"interconnect_score":157,"parallel_efficiency":158,"insufficient":206,"vram_shortfall_gb":437,"score":446,"tier":160},"RTX 5060 Ti",24,0.1072,626.05,[174],448,0.5039,false,"value",[],{"posts_count":44,"benchmarks_count":44}]