{
 "id": "qwen2.5-vl-72b",
 "name": "Qwen2.5-VL 72B",
 "repo": "Qwen/Qwen2.5-VL-72B-Instruct",
 "lab": {
  "name": "Alibaba",
  "country": "China"
 },
 "released": "2025-01-27",
 "parameters": 73.410777344,
 "activeParameters": null,
 "maxContext": 128000,
 "architecture": "80 layers · 80 full attention (8 KV × 128)",
 "page": "https://studiotvai.com/vram-requirements/qwen2-5-vl-72b",
 "calculator": "https://studiotvai.com/llm-vram-calculator?model=qwen2.5-vl-72b",
 "weightsGiB": {
  "bf16": 136.74,
  "fp8": 70.69,
  "q4_k_m": 43.73
 },
 "kvCacheGiBPerRequest": [
  {
   "context": 8192,
   "bf16": 2.5,
   "fp8": 1.25
  },
  {
   "context": 32768,
   "bf16": 10,
   "fp8": 5
  }
 ],
 "recurrentStateGiBPerRequest": 0,
 "gpusNeeded": {
  "scenario": "one request of 8192 tokens, BF16 KV cache; vLLM on data-center GPUs, llama.cpp elsewhere",
  "gpus": [
   {
    "gpu": "rtx-4090",
    "name": "GeForce RTX 4090",
    "memoryGB": 24,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 7,
     "fp8": 4,
     "q4_k_m": 3
    }
   },
   {
    "gpu": "rtx-5090",
    "name": "GeForce RTX 5090",
    "memoryGB": 32,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 5,
     "fp8": 3,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "rtx-pro-6000",
    "name": "RTX PRO 6000 Blackwell",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 2,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "mac-m5-max-128",
    "name": "Mac M5 Max 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 2,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "strix-halo",
    "name": "Ryzen AI Max+ 395 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 2,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "dgx-spark",
    "name": "DGX Spark (GB10)",
    "memoryGB": 128,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 2,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "h100",
    "name": "NVIDIA H100",
    "memoryGB": 80,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 3,
     "fp8": 2,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "h200",
    "name": "NVIDIA H200",
    "memoryGB": 141,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 2,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "b200",
    "name": "NVIDIA B200 (HGX)",
    "memoryGB": 180,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 1,
     "fp8": 1,
     "q4_k_m": 1
    }
   }
  ]
 },
 "ggufFiles": {
  "repo": "bartowski/Qwen_Qwen2.5-VL-72B-Instruct-GGUF",
  "gib": {
   "IQ1_M": 22.11,
   "IQ2_M": 27.32,
   "IQ2_S": 26.02,
   "IQ2_XS": 25.2,
   "IQ2_XXS": 23.74,
   "IQ3_M": 33.07,
   "IQ3_XS": 30.59,
   "IQ3_XXS": 29.66,
   "IQ4_NL": 38.48,
   "IQ4_XS": 36.98,
   "Q2_K": 27.76,
   "Q2_K_L": 28.9,
   "Q3_K_L": 36.79,
   "Q3_K_M": 35.11,
   "Q3_K_S": 32.12,
   "Q3_K_XL": 37.81,
   "Q4_0": 38.54,
   "Q4_1": 42.56,
   "Q4_K_L": 45.02,
   "Q4_K_M": 44.16,
   "Q4_K_S": 40.88,
   "Q5_K_M": 50.71,
   "Q5_K_S": 47.85,
   "Q6_K": 59.93,
   "Q8_0": 71.96
  }
 },
 "runsWith": [
  "vLLM",
  "SGLang",
  "llama.cpp (Ollama, LM Studio)",
  "MLX (Mac)"
 ],
 "verified": {
  "gguf": {
   "day": "2026-11-06",
   "repo": "bartowski/Qwen_Qwen2.5-VL-72B-Instruct-GGUF",
   "kv": "match",
   "ours": 327680,
   "theirs": 327680
  }
 },
 "updated": "2026-11-06T21:02:26.190Z"
}