{
 "id": "hy3",
 "name": "Hy3",
 "repo": "tencent/Hy3",
 "lab": {
  "name": "Tencent",
  "country": "China"
 },
 "released": "2026-07-02",
 "parameters": 298.786155776,
 "activeParameters": 20,
 "maxContext": 262144,
 "architecture": "80 layers · 80 full attention (8 KV × 128)",
 "page": "https://studiotvai.com/vram-requirements/hy3",
 "calculator": "https://studiotvai.com/llm-vram-calculator?model=hy3",
 "weightsGiB": {
  "bf16": 549.54,
  "fp8": 275.69,
  "q4_k_m": 165.99
 },
 "kvCacheGiBPerRequest": [
  {
   "context": 8192,
   "bf16": 2.5,
   "fp8": 1.25
  },
  {
   "context": 32768,
   "bf16": 10,
   "fp8": 5
  },
  {
   "context": 131072,
   "bf16": 40,
   "fp8": 20
  },
  {
   "context": 262144,
   "bf16": 80,
   "fp8": 40
  }
 ],
 "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": 32,
     "fp8": 16,
     "q4_k_m": 8
    }
   },
   {
    "gpu": "rtx-5090",
    "name": "GeForce RTX 5090",
    "memoryGB": 32,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 24,
     "fp8": 16,
     "q4_k_m": 6
    }
   },
   {
    "gpu": "rtx-pro-6000",
    "name": "RTX PRO 6000 Blackwell",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 6,
     "fp8": 3,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "mac-m5-max-128",
    "name": "Mac M5 Max 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 6,
     "fp8": 3,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "strix-halo",
    "name": "Ryzen AI Max+ 395 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 6,
     "fp8": 3,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "dgx-spark",
    "name": "DGX Spark (GB10)",
    "memoryGB": 128,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 5,
     "fp8": 3,
     "q4_k_m": 2
    }
   },
   {
    "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": 5,
     "fp8": 3,
     "q4_k_m": 2
    }
   },
   {
    "gpu": "b200",
    "name": "NVIDIA B200 (HGX)",
    "memoryGB": 180,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 4,
     "fp8": 2,
     "q4_k_m": 2
    }
   }
  ]
 },
 "ggufFiles": {
  "repo": "bartowski/Hy3-GGUF",
  "gib": {
   "IQ1_M": 66.22,
   "IQ1_S": 59.65,
   "IQ2_M": 95.08,
   "IQ2_S": 86.43,
   "IQ2_XS": 84.81,
   "IQ2_XXS": 76.47,
   "IQ3_M": 133.8,
   "IQ3_XS": 128,
   "IQ3_XXS": 117.43,
   "IQ4_NL": 158.14,
   "IQ4_XS": 149.73,
   "Q2_K": 99.36,
   "Q2_K_L": 99.81,
   "Q3_K_L": 133.25,
   "Q3_K_M": 128.1,
   "Q3_K_S": 122.26,
   "Q3_K_XL": 133.65,
   "Q4_0": 158.62,
   "Q4_1": 174.9,
   "Q4_K_L": 169.99,
   "Q4_K_M": 169.65,
   "Q4_K_S": 163.41,
   "Q5_K_M": 198.22,
   "Q5_K_S": 191.88,
   "Q6_K": 239.58,
   "Q8_0": 295.84
  }
 },
 "runsWith": [
  "vLLM",
  "SGLang",
  "llama.cpp (Ollama, LM Studio)",
  "MLX (Mac)"
 ],
 "verified": {
  "gguf": {
   "day": "2026-10-17",
   "repo": "bartowski/Hy3-GGUF",
   "kv": "match",
   "ours": 327680,
   "theirs": 327680
  }
 },
 "updated": "2026-10-17T00:05:06.000Z"
}