{
 "id": "qwen3.8-27b-dspine-b16",
 "name": "Qwen3.8 27B-DSpine-b16",
 "repo": "tencent/Qwen3.8-27B-DSpine-b16",
 "lab": {
  "name": "Tencent",
  "country": "China"
 },
 "released": "2026-11-07",
 "parameters": 2.025912064,
 "activeParameters": null,
 "maxContext": 262144,
 "architecture": "5 layers · 5 sliding-window layers (8192 tokens, 8 KV × 128)",
 "page": "https://studiotvai.com/vram-requirements/qwen3-8-27b-dspine-b16",
 "calculator": "https://studiotvai.com/llm-vram-calculator?model=qwen3.8-27b-dspine-b16",
 "weightsGiB": {
  "bf16": 3.77,
  "fp8": 2.83,
  "q4_k_m": 1.4
 },
 "kvCacheGiBPerRequest": [
  {
   "context": 8192,
   "bf16": 0.156,
   "fp8": 0.078
  },
  {
   "context": 32768,
   "bf16": 0.156,
   "fp8": 0.078
  },
  {
   "context": 131072,
   "bf16": 0.156,
   "fp8": 0.078
  },
  {
   "context": 262144,
   "bf16": 0.156,
   "fp8": 0.078
  }
 ],
 "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": 1,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "rtx-5090",
    "name": "GeForce RTX 5090",
    "memoryGB": 32,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 1,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "rtx-pro-6000",
    "name": "RTX PRO 6000 Blackwell",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 1,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "mac-m5-max-128",
    "name": "Mac M5 Max 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 1,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "strix-halo",
    "name": "Ryzen AI Max+ 395 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 1,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "dgx-spark",
    "name": "DGX Spark (GB10)",
    "memoryGB": 128,
    "engine": "llama.cpp",
    "byFormat": {
     "bf16": 1,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "h100",
    "name": "NVIDIA H100",
    "memoryGB": 80,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 1,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "h200",
    "name": "NVIDIA H200",
    "memoryGB": 141,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 1,
     "fp8": 1,
     "q4_k_m": 1
    }
   },
   {
    "gpu": "b200",
    "name": "NVIDIA B200 (HGX)",
    "memoryGB": 180,
    "engine": "vLLM",
    "byFormat": {
     "bf16": 1,
     "fp8": 1,
     "q4_k_m": 1
    }
   }
  ]
 },
 "ggufFiles": {
  "repo": "aj9o9/Qwen3.8-27B-DSpine-b16-GGUF",
  "gib": {
   "BF16": 3.78,
   "Q4_K_M": 1.17,
   "Q8_0": 2.05
  }
 },
 "runsWith": [
  "llama.cpp (Ollama, LM Studio)",
  "MLX (Mac)"
 ],
 "verified": null,
 "updated": "2026-11-22T00:05:05.824Z"
}
