{
 "id": "gpt-oss-120b",
 "name": "gpt-oss 120B",
 "repo": "openai/gpt-oss-120b",
 "lab": {
  "name": "OpenAI",
  "country": "United States"
 },
 "released": "2025-08-04",
 "parameters": 116.83,
 "activeParameters": 5.1,
 "maxContext": 131072,
 "architecture": "MoE 128 experts · alternating 128-token sliding window / full attention (8 KV × 64)",
 "page": "https://studiotvai.com/vram-requirements/gpt-oss-120b",
 "calculator": "https://studiotvai.com/llm-vram-calculator?model=gpt-oss-120b",
 "weightsGiB": {
  "native": 60.77,
  "bf16": 217.61,
  "fp8": 109.89,
  "q4_k_m": 58.27
 },
 "kvCacheGiBPerRequest": [
  {
   "context": 8192,
   "bf16": 0.286,
   "fp8": 0.143
  },
  {
   "context": 32768,
   "bf16": 1.129,
   "fp8": 0.565
  },
  {
   "context": 131072,
   "bf16": 4.504,
   "fp8": 2.252
  }
 ],
 "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": 3,
     "bf16": 16,
     "fp8": 6,
     "q4_k_m": 3
    }
   },
   {
    "gpu": "rtx-5090",
    "name": "GeForce RTX 5090",
    "memoryGB": 32,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 3,
     "bf16": 8,
     "fp8": 4,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "rtx-pro-6000",
    "name": "RTX PRO 6000 Blackwell",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 1,
     "bf16": 3,
     "fp8": 2,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "mac-m5-max-128",
    "name": "Mac M5 Max 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 1,
     "bf16": 3,
     "fp8": 2,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "h100",
    "name": "NVIDIA H100",
    "memoryGB": 80,
    "engine": "vLLM",
    "byFormat": {
     "native": 1,
     "bf16": 4,
     "fp8": 2,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "h200",
    "name": "NVIDIA H200",
    "memoryGB": 141,
    "engine": "vLLM",
    "byFormat": {
     "native": 1,
     "bf16": 2,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "b200",
    "name": "NVIDIA B200 (HGX)",
    "memoryGB": 180,
    "engine": "vLLM",
    "byFormat": {
     "native": 1,
     "bf16": 2,
     "fp8": 1,
     "q4_k_m": 1
    }
   }
  ]
 },
 "ggufFiles": {
  "repo": "unsloth/gpt-oss-120b-GGUF",
  "gib": {
   "Q2_K": 58.27,
   "Q2_K_L": 58.54,
   "Q3_K_M": 58.33,
   "Q3_K_S": 58.27,
   "Q4_0": 58.32,
   "Q4_1": 58.41,
   "Q4_K_M": 58.46,
   "Q4_K_S": 58.45,
   "Q5_K_M": 58.57,
   "Q5_K_S": 58.56,
   "Q6_K": 58.94,
   "Q8_0": 59.03,
   "UD-Q4_K_XL": 58.69,
   "UD-Q6_K_XL": 58.94,
   "UD-Q8_K_XL": 60.05,
   "F16": 60.88
  }
 },
 "runsWith": [
  "vLLM",
  "SGLang",
  "llama.cpp (Ollama, LM Studio)",
  "MLX (Mac)"
 ],
 "verified": null,
 "updated": "2026-10-10T15:05:26.095Z"
}