{
 "id": "qwen2.5-vl-32b",
 "name": "Qwen2.5-VL 32B",
 "repo": "Qwen/Qwen2.5-VL-32B-Instruct",
 "lab": {
  "name": "Alibaba",
  "country": "China"
 },
 "released": "2025-03-21",
 "parameters": 33.452718336,
 "activeParameters": null,
 "maxContext": 128000,
 "architecture": "64 layers · 64 full attention (8 KV × 128)",
 "page": "https://studiotvai.com/vram-requirements/qwen2-5-vl-32b",
 "calculator": "https://studiotvai.com/llm-vram-calculator?model=qwen2.5-vl-32b",
 "weightsGiB": {
  "bf16": 62.31,
  "fp8": 32.61,
  "q4_k_m": 18.58
 },
 "kvCacheGiBPerRequest": [
  {
   "context": 8192,
   "bf16": 2,
   "fp8": 1
  },
  {
   "context": 32768,
   "bf16": 8,
   "fp8": 4
  }
 ],
 "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": 4,
     "fp8": 2,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "rtx-5090",
    "name": "GeForce RTX 5090",
    "memoryGB": 32,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 3,
     "fp8": 2,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "rtx-pro-6000",
    "name": "RTX PRO 6000 Blackwell",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 1,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "mac-m5-max-128",
    "name": "Mac M5 Max 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 1,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "h100",
    "name": "NVIDIA H100",
    "memoryGB": 80,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 1,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "h200",
    "name": "NVIDIA H200",
    "memoryGB": 141,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 1,
     "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-32B-Instruct-GGUF",
  "gib": {
   "IQ2_M": 10.49,
   "IQ2_S": 9.67,
   "IQ2_XS": 9.27,
   "IQ2_XXS": 8.41,
   "IQ3_M": 13.79,
   "IQ3_XS": 12.76,
   "IQ3_XXS": 11.96,
   "IQ4_NL": 17.4,
   "IQ4_XS": 16.48,
   "Q2_K": 11.47,
   "Q2_K_L": 12.18,
   "Q3_K_L": 16.06,
   "Q3_K_M": 14.84,
   "Q3_K_S": 13.4,
   "Q3_K_XL": 16.7,
   "Q4_0": 17.43,
   "Q4_1": 19.22,
   "Q4_K_L": 19.03,
   "Q4_K_M": 18.49,
   "Q4_K_S": 17.49,
   "Q5_K_L": 22.11,
   "Q5_K_M": 21.66,
   "Q5_K_S": 21.08,
   "Q6_K": 25.04,
   "Q6_K_L": 25.39,
   "Q8_0": 32.43,
   "BF16": 61.04
  }
 },
 "runsWith": [
  "vLLM",
  "SGLang",
  "llama.cpp (Ollama, LM Studio)",
  "MLX (Mac)"
 ],
 "verified": {
  "gguf": {
   "day": "2026-10-14",
   "repo": "bartowski/Qwen_Qwen2.5-VL-32B-Instruct-GGUF",
   "kv": "match",
   "ours": 262144,
   "theirs": 262144
  }
 },
 "updated": "2026-10-14T00:07:05.854Z"
}