{
 "id": "glm-4.5",
 "name": "GLM-4.5",
 "repo": "zai-org/GLM-4.5",
 "lab": {
  "name": "Z.ai",
  "country": "China"
 },
 "released": "2025-07-20",
 "parameters": 358.337791296,
 "activeParameters": 33,
 "maxContext": 131072,
 "architecture": "92 layers · 92 full attention (8 KV × 128)",
 "page": "https://studiotvai.com/vram-requirements/glm-4-5",
 "calculator": "https://studiotvai.com/llm-vram-calculator?model=glm-4.5",
 "weightsGiB": {
  "bf16": 657.14,
  "fp8": 330.01,
  "q4_k_m": 198.53
 },
 "kvCacheGiBPerRequest": [
  {
   "context": 8192,
   "bf16": 2.875,
   "fp8": 1.438
  },
  {
   "context": 32768,
   "bf16": 11.5,
   "fp8": 5.75
  },
  {
   "context": 131072,
   "bf16": 46,
   "fp8": 23
  }
 ],
 "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": 48,
     "fp8": 24,
     "q4_k_m": 16
    }
   },
   {
    "gpu": "rtx-5090",
    "name": "GeForce RTX 5090",
    "memoryGB": 32,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 24,
     "fp8": 16,
     "q4_k_m": 7
    }
   },
   {
    "gpu": "rtx-pro-6000",
    "name": "RTX PRO 6000 Blackwell",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 8,
     "fp8": 4,
     "q4_k_m": 3
    }
   },
   {
    "gpu": "mac-m5-max-128",
    "name": "Mac M5 Max 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 8,
     "fp8": 4,
     "q4_k_m": 3
    }
   },
   {
    "gpu": "h100",
    "name": "NVIDIA H100",
    "memoryGB": 80,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 16,
     "fp8": 5,
     "q4_k_m": 3
    }
   },
   {
    "gpu": "h200",
    "name": "NVIDIA H200",
    "memoryGB": 141,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 6,
     "fp8": 3,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "b200",
    "name": "NVIDIA B200 (HGX)",
    "memoryGB": 180,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 5,
     "fp8": 3,
     "q4_k_m": 2
    }
   }
  ]
 },
 "ggufFiles": {
  "repo": "unsloth/GLM-4.5-GGUF",
  "gib": {
   "BF16": 667.61,
   "UD-TQ1_0": 78.54,
   "IQ4_NL": 188.55,
   "IQ4_XS": 178.41,
   "Q2_K": 122.15,
   "Q2_K_L": 122.32,
   "Q3_K_M": 159.51,
   "Q3_K_S": 144.38,
   "Q4_0": 189.09,
   "Q4_1": 209.18,
   "Q4_K_M": 201.59,
   "Q4_K_S": 189.69,
   "Q5_K_M": 236.79,
   "Q5_K_S": 230.04,
   "Q6_K": 274.16,
   "Q8_0": 354.8,
   "UD-IQ1_M": 100.32,
   "UD-IQ1_S": 90.38,
   "UD-IQ2_M": 114.03,
   "UD-IQ2_XXS": 107.95,
   "UD-IQ3_XXS": 135.21,
   "UD-Q2_K_XL": 125.91,
   "UD-Q3_K_XL": 147.84,
   "UD-Q4_K_XL": 190.37,
   "UD-Q5_K_XL": 235.74,
   "UD-Q6_K_XL": 279.61,
   "UD-Q8_K_XL": 364.59
  }
 },
 "runsWith": [
  "vLLM",
  "SGLang",
  "llama.cpp (Ollama, LM Studio)",
  "MLX (Mac)"
 ],
 "verified": {
  "gguf": {
   "day": "2026-10-14",
   "repo": "unsloth/GLM-4.5-GGUF",
   "kv": "match",
   "ours": 376832,
   "theirs": 376832
  }
 },
 "updated": "2026-10-14T00:07:05.854Z"
}