{
 "id": "minicpm-v-4.7-35b-a3b",
 "name": "MiniCPM-V-4.7 35B-A3B",
 "repo": "openbmb/MiniCPM-V-4.7-35B-A3B",
 "lab": {
  "name": "OpenBMB",
  "country": "China"
 },
 "released": "2026-10-07",
 "parameters": 35.212875824,
 "activeParameters": 3,
 "maxContext": 262144,
 "architecture": "40 layers · 10 full attention (2 KV × 256) · 30 linear attention",
 "page": "https://studiotvai.com/vram-requirements/minicpm-v-4-7-35b-a3b",
 "calculator": "https://studiotvai.com/llm-vram-calculator?model=minicpm-v-4.7-35b-a3b",
 "weightsGiB": {
  "bf16": 65.59,
  "fp8": 33.74,
  "q4_k_m": 19.61
 },
 "kvCacheGiBPerRequest": [
  {
   "context": 8192,
   "bf16": 0.156,
   "fp8": 0.078
  },
  {
   "context": 32768,
   "bf16": 0.625,
   "fp8": 0.313
  },
  {
   "context": 131072,
   "bf16": 2.5,
   "fp8": 1.25
  },
  {
   "context": 262144,
   "bf16": 5,
   "fp8": 2.5
  }
 ],
 "recurrentStateGiBPerRequest": 0.06,
 "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": "openbmb/MiniCPM-V-4.7-35B-A3B-gguf",
  "gib": {
   "F16": 64.61,
   "Q4_0": 18.36,
   "Q4_1": 20.35,
   "Q4_K_M": 19.71,
   "Q4_K_S": 18.52,
   "Q5_0": 22.33,
   "Q5_1": 24.32,
   "Q5_K_M": 23.03,
   "Q5_K_S": 22.33,
   "Q6_K": 26.56,
   "Q8_0": 34.37
  }
 },
 "runsWith": [
  "vLLM",
  "llama.cpp (Ollama, LM Studio)"
 ],
 "verified": null,
 "updated": "2026-10-12T18:02:05.777Z"
}