{
 "id": "glm-5.3-flash",
 "name": "GLM-5.3-Flash",
 "repo": "zai-org/GLM-5.3-Flash",
 "lab": {
  "name": "Z.ai",
  "country": "China"
 },
 "released": "2026-08-25",
 "parameters": 321.32,
 "activeParameters": 18,
 "maxContext": 1048576,
 "architecture": "MoE 288 experts · sparse MLA (512 latent, no RoPE) on 11 of 45 layers, indexer keys pooled by 4 · 34 Kimi Delta Attention layers",
 "page": "https://studiotvai.com/vram-requirements/glm-5-3-flash",
 "calculator": "https://studiotvai.com/llm-vram-calculator?model=glm-5.3-flash",
 "weightsGiB": {
  "native": 298.81,
  "bf16": 584.66,
  "fp8": 293.51,
  "q4_k_m": 176.31
 },
 "kvCacheGiBPerRequest": [
  {
   "context": 8192,
   "bf16": 0.089,
   "fp8": 0.058
  },
  {
   "context": 32768,
   "bf16": 0.355,
   "fp8": 0.231
  },
  {
   "context": 131072,
   "bf16": 1.419,
   "fp8": 0.925
  },
  {
   "context": 1048576,
   "bf16": 11.354,
   "fp8": 7.401
  }
 ],
 "recurrentStateGiBPerRequest": 0.1375,
 "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": 32,
     "fp8": 16,
     "q4_k_m": 16
    }
   },
   {
    "gpu": "rtx-5090",
    "name": "GeForce RTX 5090",
    "memoryGB": 32,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 16,
     "bf16": 24,
     "fp8": 16,
     "q4_k_m": 7
    }
   },
   {
    "gpu": "rtx-pro-6000",
    "name": "RTX PRO 6000 Blackwell",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 4,
     "bf16": 7,
     "fp8": 4,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "mac-m5-max-128",
    "name": "Mac M5 Max 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 4,
     "bf16": 7,
     "fp8": 4,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "h100",
    "name": "NVIDIA H100",
    "memoryGB": 80,
    "engine": "vLLM",
    "byFormat": {
     "native": 5,
     "bf16": 16,
     "fp8": 5,
     "q4_k_m": 3
    }
   },
   {
    "gpu": "h200",
    "name": "NVIDIA H200",
    "memoryGB": 141,
    "engine": "vLLM",
    "byFormat": {
     "native": 3,
     "bf16": 5,
     "fp8": 3,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "b200",
    "name": "NVIDIA B200 (HGX)",
    "memoryGB": 180,
    "engine": "vLLM",
    "byFormat": {
     "native": 2,
     "bf16": 4,
     "fp8": 2,
     "q4_k_m": 2
    }
   }
  ]
 },
 "ggufFiles": {
  "repo": "unsloth/GLM-5.3-Flash-GGUF",
  "gib": {
   "BF16": 597.57,
   "Q8_0": 317.56,
   "UD-IQ1_M": 90.88,
   "UD-IQ1_S": 86.69,
   "UD-IQ2_XXS": 94.85,
   "UD-IQ3_XXS": 112.1,
   "UD-IQ4_XS": 146.05,
   "UD-Q2_K_XL": 101.25,
   "UD-Q3_K_XL": 137.4,
   "UD-Q4_K_XL": 185.99,
   "UD-Q5_K_XL": 223.8,
   "UD-Q6_K_XL": 271.79
  }
 },
 "runsWith": [
  "vLLM",
  "SGLang",
  "llama.cpp (Ollama, LM Studio)",
  "MLX (Mac)"
 ],
 "verified": null,
 "updated": "2026-10-10T15:05:26.095Z"
}