{
 "id": "ling-3.1-flash",
 "name": "Ling-3.1-flash",
 "repo": "inclusionAI/Ling-3.1-flash",
 "lab": {
  "name": "Ant Group",
  "country": "China"
 },
 "released": "2026-10-13",
 "parameters": 572.120680768,
 "activeParameters": 24,
 "maxContext": 262144,
 "architecture": "72 layers · MLA (512 + 64 latent) on 9 · 63 Kimi Delta Attention",
 "page": "https://studiotvai.com/vram-requirements/ling-3-1-flash",
 "calculator": "https://studiotvai.com/llm-vram-calculator?model=ling-3.1-flash",
 "weightsGiB": {
  "native": 526.73,
  "bf16": 1050.43,
  "fp8": 526.42,
  "q4_k_m": 317.23
 },
 "kvCacheGiBPerRequest": [
  {
   "context": 8192,
   "bf16": 0.079,
   "fp8": 0.04
  },
  {
   "context": 32768,
   "bf16": 0.316,
   "fp8": 0.158
  },
  {
   "context": 131072,
   "bf16": 1.266,
   "fp8": 0.633
  },
  {
   "context": 262144,
   "bf16": 2.531,
   "fp8": 1.266
  }
 ],
 "recurrentStateGiBPerRequest": 0.2547,
 "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": 32,
     "bf16": 64,
     "fp8": 32,
     "q4_k_m": 16
    }
   },
   {
    "gpu": "rtx-5090",
    "name": "GeForce RTX 5090",
    "memoryGB": 32,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 24,
     "bf16": 48,
     "fp8": 24,
     "q4_k_m": 16
    }
   },
   {
    "gpu": "rtx-pro-6000",
    "name": "RTX PRO 6000 Blackwell",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 6,
     "bf16": 16,
     "fp8": 6,
     "q4_k_m": 4
    }
   },
   {
    "gpu": "mac-m5-max-128",
    "name": "Mac M5 Max 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 6,
     "bf16": 16,
     "fp8": 6,
     "q4_k_m": 4
    }
   },
   {
    "gpu": "strix-halo",
    "name": "Ryzen AI Max+ 395 128 GB",
    "memoryGB": 96,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 6,
     "bf16": 16,
     "fp8": 6,
     "q4_k_m": 4
    }
   },
   {
    "gpu": "dgx-spark",
    "name": "DGX Spark (GB10)",
    "memoryGB": 128,
    "engine": "llama.cpp",
    "byFormat": {
     "native": 5,
     "bf16": 16,
     "fp8": 5,
     "q4_k_m": 3
    }
   },
   {
    "gpu": "h100",
    "name": "NVIDIA H100",
    "memoryGB": 80,
    "engine": "vLLM",
    "byFormat": {
     "native": 8,
     "bf16": 16,
     "fp8": 8,
     "q4_k_m": 5
    }
   },
   {
    "gpu": "h200",
    "name": "NVIDIA H200",
    "memoryGB": 141,
    "engine": "vLLM",
    "byFormat": {
     "native": 5,
     "bf16": 16,
     "fp8": 5,
     "q4_k_m": 3
    }
   },
   {
    "gpu": "b200",
    "name": "NVIDIA B200 (HGX)",
    "memoryGB": 180,
    "engine": "vLLM",
    "byFormat": {
     "native": 4,
     "bf16": 7,
     "fp8": 4,
     "q4_k_m": 3
    }
   }
  ]
 },
 "ggufFiles": {
  "repo": "AtomicChat/Ling-3.1-flash-GGUF",
  "gib": {
   "IQ1_M": 149.88,
   "IQ1_S": 132.07,
   "IQ2_M": 203.1,
   "IQ2_XS": 184.61,
   "IQ2_XXS": 162.18,
   "IQ3_M": 258.04,
   "IQ3_XXS": 238.79,
   "IQ4_XS": 314.65,
   "IQ4_XXS": 290.04,
   "Q4_K_M": 330.06,
   "Q5_K_M": 372.75,
   "Q6_K": 445.15,
   "Q8_0": 559.58
  }
 },
 "runsWith": [
  "vLLM",
  "SGLang",
  "llama.cpp (Ollama, LM Studio)",
  "MLX (Mac)"
 ],
 "verified": null,
 "updated": "2026-10-16T00:03:05.821Z"
}