{
 "id": "k-exaone-2.0",
 "name": "K-EXAONE 2.0 750B-A37B",
 "repo": "LGAI-EXAONE/K-EXAONE-2.0-750B-A37B",
 "lab": {
  "name": "LG AI Research",
  "country": "South Korea"
 },
 "released": "2026-07-29",
 "parameters": 749.36,
 "activeParameters": 37,
 "maxContext": 262144,
 "architecture": "MoE 256 experts + 1 shared · 20 global layers + 58 sliding-window layers (57 × 128 tokens, 1 × 4096), GQA 8 KV × 128",
 "page": "https://studiotvai.com/vram-requirements/k-exaone-2-0",
 "calculator": "https://studiotvai.com/llm-vram-calculator?model=k-exaone-2.0",
 "weightsGiB": {
  "bf16": 1394.81,
  "fp8": 699.16,
  "q4_k_m": 421.25
 },
 "kvCacheGiBPerRequest": [
  {
   "context": 8192,
   "bf16": 0.668,
   "fp8": 0.334
  },
  {
   "context": 32768,
   "bf16": 2.543,
   "fp8": 1.272
  },
  {
   "context": 131072,
   "bf16": 10.043,
   "fp8": 5.022
  },
  {
   "context": 262144,
   "bf16": 20.043,
   "fp8": 10.022
  }
 ],
 "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": 96,
     "fp8": 48,
     "q4_k_m": 24
    }
   },
   {
    "gpu": "rtx-5090",
    "name": "GeForce RTX 5090",
    "memoryGB": 32,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 64,
     "fp8": 32,
     "q4_k_m": 16
    }
   },
   {
    "gpu": "rtx-pro-6000",
    "name": "RTX PRO 6000 Blackwell",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 16,
     "fp8": 8,
     "q4_k_m": 5
    }
   },
   {
    "gpu": "mac-m5-max-128",
    "name": "Mac M5 Max 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 16,
     "fp8": 8,
     "q4_k_m": 5
    }
   },
   {
    "gpu": "strix-halo",
    "name": "Ryzen AI Max+ 395 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 16,
     "fp8": 8,
     "q4_k_m": 5
    }
   },
   {
    "gpu": "dgx-spark",
    "name": "DGX Spark (GB10)",
    "memoryGB": 128,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 16,
     "fp8": 6,
     "q4_k_m": 4
    }
   },
   {
    "gpu": "h100",
    "name": "NVIDIA H100",
    "memoryGB": 80,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 24,
     "fp8": 16,
     "q4_k_m": 7
    }
   },
   {
    "gpu": "h200",
    "name": "NVIDIA H200",
    "memoryGB": 141,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 16,
     "fp8": 6,
     "q4_k_m": 4
    }
   },
   {
    "gpu": "b200",
    "name": "NVIDIA B200 (HGX)",
    "memoryGB": 180,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 16,
     "fp8": 5,
     "q4_k_m": 3
    }
   }
  ]
 },
 "ggufFiles": {
  "repo": "DevQuasar/NOT_WOKRKING-LGAI-EXAONE.K-EXAONE-2.0-750B-A37B-GGUF",
  "gib": {
   "Q2_K": 253.61
  }
 },
 "runsWith": [
  "vLLM",
  "SGLang",
  "llama.cpp (Ollama, LM Studio)",
  "MLX (Mac)"
 ],
 "verified": null,
 "updated": "2026-10-30T00:07:05.915Z"
}