{
 "source": "https://vramneeded.com/compare/",
 "checked": "2026-10-23",
 "units": "bytes (decimal); KV cache per token at FP16",
 "models": [
  {
   "id": "llama-3-2-3b",
   "name": "Llama 3.2 3B Instruct",
   "maker": "Meta",
   "params_total": 3212749824,
   "q4_k_m_bytes": 2019377696,
   "q8_0_bytes": 3421899296,
   "bf16_bytes": 6433687840,
   "kv_bytes_per_token_fp16": 114688,
   "context_max": 131072,
   "page": "https://vramneeded.com/compare/#row-llama-3-2-3b",
   "sources": [
    "https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct",
    "https://huggingface.co/bartowski/Llama-3.2-3B-Instruct-GGUF/tree/main",
    "https://huggingface.co/unsloth/Llama-3.2-3B-Instruct/blob/main/config.json",
    "https://ollama.com/library/llama3.2"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "llama-3-1-8b",
   "name": "Llama 3.1 8B Instruct",
   "maker": "Meta",
   "params_total": 8030261248,
   "q4_k_m_bytes": 4920739232,
   "q8_0_bytes": 8540775840,
   "bf16_bytes": 16060522496,
   "kv_bytes_per_token_fp16": 131072,
   "context_max": 131072,
   "page": "https://vramneeded.com/compare/#row-llama-3-1-8b",
   "sources": [
    "https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct",
    "https://huggingface.co/bartowski/Meta-Llama-3.1-8B-Instruct-GGUF/tree/main",
    "https://huggingface.co/unsloth/Meta-Llama-3.1-8B-Instruct/blob/main/config.json",
    "https://ollama.com/library/llama3.1"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "deepseek-r1-8b",
   "name": "DeepSeek-R1-0528-Qwen3-8B",
   "maker": "DeepSeek",
   "params_total": 8190735360,
   "q4_k_m_bytes": 5027783040,
   "q8_0_bytes": 8709517696,
   "bf16_bytes": 16388042816,
   "kv_bytes_per_token_fp16": 147456,
   "context_max": 131072,
   "page": "https://vramneeded.com/compare/#row-deepseek-r1-8b",
   "sources": [
    "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
    "https://huggingface.co/bartowski/deepseek-ai_DeepSeek-R1-0528-Qwen3-8B-GGUF/tree/main",
    "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528-Qwen3-8B/blob/main/config.json",
    "https://ollama.com/library/deepseek-r1"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "gpt-oss-20b",
   "name": "gpt-oss-20b",
   "maker": "OpenAI",
   "params_total": 20914757184,
   "q4_k_m_bytes": 11673418816,
   "q8_0_bytes": 12109566016,
   "bf16_bytes": 13792637760,
   "kv_bytes_per_token_fp16": 24576,
   "context_max": 131072,
   "page": "https://vramneeded.com/compare/#row-gpt-oss-20b",
   "sources": [
    "https://huggingface.co/openai/gpt-oss-20b",
    "https://huggingface.co/bartowski/openai_gpt-oss-20b-GGUF/tree/main",
    "https://huggingface.co/openai/gpt-oss-20b/blob/main/config.json",
    "https://ollama.com/library/gpt-oss"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "qwen3-14b",
   "name": "Qwen3 14B",
   "maker": "Alibaba Qwen",
   "params_total": 14768307200,
   "q4_k_m_bytes": 9001753632,
   "q8_0_bytes": 15698534432,
   "bf16_bytes": 29543423776,
   "kv_bytes_per_token_fp16": 163840,
   "context_max": null,
   "page": "https://vramneeded.com/models/qwen3-14b/",
   "sources": [
    "https://huggingface.co/Qwen/Qwen3-14B",
    "https://huggingface.co/bartowski/Qwen_Qwen3-14B-GGUF/tree/main",
    "https://huggingface.co/Qwen/Qwen3-14B/blob/main/config.json",
    "https://ollama.com/library/qwen3/tags"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "gemma-4-26b-a4b",
   "name": "Gemma 4 26B A4B",
   "maker": "Google",
   "params_total": 25805936206,
   "q4_k_m_bytes": 17035039872,
   "q8_0_bytes": 26859861120,
   "bf16_bytes": 50505135424,
   "kv_bytes_per_token_fp16": null,
   "context_max": 262144,
   "page": "https://vramneeded.com/compare/#row-gemma-4-26b-a4b",
   "sources": [
    "https://huggingface.co/google/gemma-4-26B-A4B-it",
    "https://huggingface.co/bartowski/google_gemma-4-26B-A4B-it-GGUF/tree/main",
    "https://huggingface.co/google/gemma-4-26B-A4B-it/blob/main/config.json",
    "https://ollama.com/library/gemma4"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "qwen3-8-27b",
   "name": "Qwen3.8 27B",
   "maker": "Alibaba Qwen",
   "params_total": 27781427952,
   "q4_k_m_bytes": 17442399968,
   "q8_0_bytes": 29116388960,
   "bf16_bytes": 54657734144,
   "kv_bytes_per_token_fp16": 65536,
   "context_max": 262144,
   "page": "https://vramneeded.com/models/qwen3-8-27b/",
   "sources": [
    "https://huggingface.co/Qwen/Qwen3.8-27B",
    "https://huggingface.co/bartowski/Qwen3.8-27B-GGUF/tree/main",
    "https://huggingface.co/Qwen/Qwen3.8-27B/blob/main/config.json",
    "https://ollama.com/library/qwen3.8"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "gemma-4-31b",
   "name": "Gemma 4 31B",
   "maker": "Google",
   "params_total": 31273088876,
   "q4_k_m_bytes": 19598489952,
   "q8_0_bytes": 32635677024,
   "bf16_bytes": 61413189920,
   "kv_bytes_per_token_fp16": null,
   "context_max": 262144,
   "page": "https://vramneeded.com/compare/#row-gemma-4-31b",
   "sources": [
    "https://huggingface.co/google/gemma-4-31B-it",
    "https://huggingface.co/bartowski/google_gemma-4-31B-it-GGUF/tree/main",
    "https://huggingface.co/google/gemma-4-31B-it/blob/main/config.json",
    "https://ollama.com/library/gemma4"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "llama-3-1-70b",
   "name": "Llama 3.1 70B Instruct",
   "maker": "Meta",
   "params_total": 70553706496,
   "q4_k_m_bytes": 42520398400,
   "q8_0_bytes": 74975054592,
   "bf16_bytes": 141107412992,
   "kv_bytes_per_token_fp16": 327680,
   "context_max": 131072,
   "page": "https://vramneeded.com/models/llama-3-1-70b/",
   "sources": [
    "https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct",
    "https://huggingface.co/bartowski/Meta-Llama-3.1-70B-Instruct-GGUF/tree/main",
    "https://huggingface.co/unsloth/Meta-Llama-3.1-70B-Instruct/blob/main/config.json",
    "https://ollama.com/library/llama3.1"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "gpt-oss-120b",
   "name": "gpt-oss-120b",
   "maker": "OpenAI",
   "params_total": 116829156672,
   "q4_k_m_bytes": 62841713344,
   "q8_0_bytes": 63387346368,
   "bf16_bytes": 65369016768,
   "kv_bytes_per_token_fp16": 36864,
   "context_max": 131072,
   "page": "https://vramneeded.com/compare/#row-gpt-oss-120b",
   "sources": [
    "https://huggingface.co/openai/gpt-oss-120b",
    "https://huggingface.co/bartowski/openai_gpt-oss-120b-GGUF/tree/main",
    "https://huggingface.co/openai/gpt-oss-120b/blob/main/config.json",
    "https://ollama.com/library/gpt-oss"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "qwen3-8-flash-next",
   "name": "Qwen3.8 Flash Next",
   "maker": "Alibaba Qwen",
   "params_total": 179999981459,
   "q4_k_m_bytes": 119599030080,
   "q8_0_bytes": 188260769536,
   "bf16_bytes": 354029929568,
   "kv_bytes_per_token_fp16": null,
   "context_max": 262144,
   "page": "https://vramneeded.com/compare/#row-qwen3-8-flash-next",
   "sources": [
    "https://huggingface.co/Qwen/Qwen3.8-Flash-Next",
    "https://huggingface.co/bartowski/Qwen3.8-Flash-Next-GGUF/tree/main",
    "https://huggingface.co/Qwen/Qwen3.8-Flash-Next/blob/main/config.json"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "mistral-7b",
   "name": "Mistral 7B Instruct v0.3",
   "maker": "Mistral AI",
   "params_total": 7248023552,
   "q4_k_m_bytes": 4372812000,
   "q8_0_bytes": 7702565088,
   "bf16_bytes": 14496047104,
   "kv_bytes_per_token_fp16": 131072,
   "context_max": 32768,
   "page": null,
   "sources": [
    "https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3",
    "https://huggingface.co/bartowski/Mistral-7B-Instruct-v0.3-GGUF"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "mixtral-8x7b",
   "name": "Mixtral 8x7B Instruct v0.1",
   "maker": "Mistral AI",
   "params_total": 46702792704,
   "q4_k_m_bytes": 26441533376,
   "q8_0_bytes": 49624262592,
   "bf16_bytes": 93405585408,
   "kv_bytes_per_token_fp16": 131072,
   "context_max": 32768,
   "page": null,
   "sources": [
    "https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1",
    "https://huggingface.co/TheBloke/Mixtral-8x7B-Instruct-v0.1-GGUF"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "qwen2-5-7b",
   "name": "Qwen2.5 7B Instruct",
   "maker": "Alibaba Qwen",
   "params_total": 7615616512,
   "q4_k_m_bytes": 4683074240,
   "q8_0_bytes": 8098525888,
   "bf16_bytes": 15237853600,
   "kv_bytes_per_token_fp16": 57344,
   "context_max": 131072,
   "page": null,
   "sources": [
    "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct",
    "https://huggingface.co/bartowski/Qwen2.5-7B-Instruct-GGUF"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "qwen2-5-14b",
   "name": "Qwen2.5 14B Instruct",
   "maker": "Alibaba Qwen",
   "params_total": 14770033664,
   "q4_k_m_bytes": 8988110976,
   "q8_0_bytes": 15701598336,
   "bf16_bytes": 29547716448,
   "kv_bytes_per_token_fp16": 196608,
   "context_max": 131072,
   "page": null,
   "sources": [
    "https://huggingface.co/Qwen/Qwen2.5-14B-Instruct",
    "https://huggingface.co/bartowski/Qwen2.5-14B-Instruct-GGUF"
   ],
   "checked": "2026-10-23"
  },
  {
   "id": "qwen2-5-32b",
   "name": "Qwen2.5 32B Instruct",
   "maker": "Alibaba Qwen",
   "params_total": 32763876352,
   "q4_k_m_bytes": 19851336576,
   "q8_0_bytes": 34820885376,
   "bf16_bytes": 65535970112,
   "kv_bytes_per_token_fp16": 262144,
   "context_max": 131072,
   "page": null,
   "sources": [
    "https://huggingface.co/Qwen/Qwen2.5-32B-Instruct",
    "https://huggingface.co/bartowski/Qwen2.5-32B-Instruct-GGUF"
   ],
   "checked": "2026-10-23"
  }
 ]
}