{
 "id": "minimax-m2.5",
 "name": "MiniMax-M2.5",
 "repo": "MiniMaxAI/MiniMax-M2.5",
 "lab": {
  "name": "MiniMax",
  "country": "China"
 },
 "released": "2026-02-12",
 "parameters": 228.703644928,
 "activeParameters": 11,
 "maxContext": 196608,
 "architecture": "62 layers · 62 full attention (8 KV × 128)",
 "page": "https://studiotvai.com/vram-requirements/minimax-m2-5",
 "calculator": "https://studiotvai.com/llm-vram-calculator?model=minimax-m2.5",
 "weightsGiB": {
  "native": 214.33,
  "bf16": 425.99,
  "fp8": 214.14,
  "q4_k_m": 128.72
 },
 "kvCacheGiBPerRequest": [
  {
   "context": 8192,
   "bf16": 1.938,
   "fp8": 0.969
  },
  {
   "context": 32768,
   "bf16": 7.75,
   "fp8": 3.875
  },
  {
   "context": 131072,
   "bf16": 31,
   "fp8": 15.5
  },
  {
   "context": 196608,
   "bf16": 46.5,
   "fp8": 23.25
  }
 ],
 "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": {
     "native": 16,
     "bf16": 24,
     "fp8": 16,
     "q4_k_m": 7
    }
   },
   {
    "gpu": "rtx-5090",
    "name": "GeForce RTX 5090",
    "memoryGB": 32,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 8,
     "bf16": 16,
     "fp8": 8,
     "q4_k_m": 5
    }
   },
   {
    "gpu": "rtx-pro-6000",
    "name": "RTX PRO 6000 Blackwell",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 3,
     "bf16": 5,
     "fp8": 3,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "mac-m5-max-128",
    "name": "Mac M5 Max 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 3,
     "bf16": 5,
     "fp8": 3,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "h100",
    "name": "NVIDIA H100",
    "memoryGB": 80,
    "engine": "vLLM",
    "byFormat": {
     "native": 4,
     "bf16": 7,
     "fp8": 4,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "h200",
    "name": "NVIDIA H200",
    "memoryGB": 141,
    "engine": "vLLM",
    "byFormat": {
     "native": 2,
     "bf16": 4,
     "fp8": 2,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "b200",
    "name": "NVIDIA B200 (HGX)",
    "memoryGB": 180,
    "engine": "vLLM",
    "byFormat": {
     "native": 2,
     "bf16": 3,
     "fp8": 2,
     "q4_k_m": 1
    }
   }
  ]
 },
 "ggufFiles": {
  "repo": "unsloth/MiniMax-M2.5-GGUF",
  "gib": {
   "BF16": 426.07,
   "IQ4_NL": 120.14,
   "IQ4_XS": 113.53,
   "MXFP4_MOE": 115.27,
   "UD-TQ1_0": 51.94,
   "Q2_K": 77.58,
   "Q2_K_L": 77.72,
   "Q3_K_M": 101.77,
   "Q3_K_S": 91.91,
   "Q4_0": 120.61,
   "Q4_1": 133.39,
   "Q4_K_M": 128.84,
   "Q4_K_S": 121.1,
   "Q5_K_M": 151.16,
   "Q5_K_S": 146.66,
   "Q6_K": 174.87,
   "Q8_0": 226.44,
   "UD-IQ1_M": 63.74,
   "UD-IQ1_S": 58.86,
   "UD-IQ2_M": 72.83,
   "UD-IQ2_XXS": 69.03,
   "UD-IQ3_XXS": 86.91,
   "UD-Q2_K_XL": 80.02,
   "UD-Q3_K_XL": 94.34,
   "UD-Q4_K_XL": 122.3,
   "UD-Q5_K_XL": 150.66,
   "UD-Q6_K_XL": 180.95,
   "UD-Q8_K_XL": 243.43
  }
 },
 "runsWith": [
  "vLLM",
  "SGLang",
  "llama.cpp (Ollama, LM Studio)",
  "MLX (Mac)"
 ],
 "verified": {
  "gguf": {
   "day": "2026-10-11",
   "repo": "unsloth/MiniMax-M2.5-GGUF",
   "kv": "match",
   "ours": 253952,
   "theirs": 253952
  }
 },
 "updated": "2026-10-11T19:08:05.666Z"
}