{
  "type": "tweet",
  "id": "2046935822096855379",
  "url": "https://x.com/gkisokay/status/2046935822096855379",
  "twitterUrl": "https://twitter.com/gkisokay/status/2046935822096855379",
  "text": "The Local LLM Cheat Sheet for your 32GB RAM device\n\nI was asked to put together a practical lineup of local models that fit comfortably on a 32GB machine.\n\nAt this tier, you start getting access to real flagship-class local models, plus a growing number of custom quants. But for most people, these are the core models worth knowing first.\n\nFlagship Models\n\nQwen3.5 27B / GGUF / Q6_K_M\nThe best overall 32GB flagship. General chat, writing, research, and agent workflows. Great if you want one model that can handle almost everything well.\n\nQwen3.6-35B-A3B / GGUF / UD-Q4_K_M\nBest MoE flagship. Stronger for coding, reasoning, and tool use than most smaller generalists.\n\nGemma 4 31B / GGUF / Q6_K_M\nDense premium model. Writing, analysis, reasoning, and high-end local chat. Heavier than the MoE options, but excellent when quality matters more than speed.\n\nModels for Fast Flagship Use\n\nGemma 4 26B A4B / GGUF / Q6_K_M\nGreat balance of speed and quality for general assistant work, coding, agent tasks, and research. This is one of the best 32GB picks if you want something that feels high-end without dragging.\n\nDeepSeek-R1 Distill Qwen 32B / GGUF / Q4_K_M\nOffline reasoning engine. Best for math, logic, deliberate analysis, and step-by-step problem solving.\n\nMistral Small 24B / GGUF / Q6_K_M\nTool-calling specialist. Strong for assistants, chat workflows, local business tasks, and function calling. Available for 24GB machines.\n\nModels for Companion Use\n\nQwen3.5 9B / GGUF / Q6_K_M\nBest sidekick. Fast drafts, search loops, cheap retries, and secondary agent work. Even on a 32GB machine, you still want a smaller model around for support tasks.\n\nLlama 3.1 8B / GGUF / Q6_K_M\nLong-context companion. RAG, doc ingestion, codebase chat, and long prompts. The output quality is not the sharpest anymore, but it is still useful when needing simple tasks fast.\n\nFrom what my community tells me, the best single models are Qwen3.5 27B or Gemma 4 31B.\n\nFor two models, the strongest general pairing is Qwen3.5 27B + Qwen3.5 9B.\n\nIf you are more code-heavy, Qwen3.6-35B-A3B + Llama 3.1 8B.\n\nLet me know what models you are running on 32GB, and which ones have actually been worth the RAM.",
  "source": "Twitter for iPhone",
  "retweetCount": 350,
  "replyCount": 85,
  "likeCount": 2139,
  "quoteCount": 13,
  "viewCount": 308084,
  "createdAt": "Wed Apr 22 12:55:04 +0000 2026",
  "lang": "en",
  "bookmarkCount": 2722,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2046935822096855379",
  "displayTextRange": [
    0,
    279
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "gkisokay",
    "url": "https://x.com/gkisokay",
    "twitterUrl": "https://twitter.com/gkisokay",
    "id": "629466946",
    "name": "Graeme",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/1988420071824470016/nrXOtFnM_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/629466946/1762782644",
    "description": "",
    "location": "",
    "followers": 25399,
    "following": 3681,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Sat Jul 07 16:36:04 +0000 2012",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 26300,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 2275,
    "statusesCount": 22451,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2037902655016804496"
    ],
    "profile_bio": {
      "description": "AI agent enjoyer | Founder @amplifi_now | Helping businesses scale with AI",
      "entities": {
        "description": {
          "user_mentions": [
            {
              "id_str": "",
              "indices": [
                27,
                39
              ],
              "name": "",
              "screen_name": "amplifi_now"
            }
          ]
        },
        "url": {
          "urls": [
            {
              "display_url": "gkisokay.com",
              "expanded_url": "http://gkisokay.com",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/VGu1PsM6b0"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "allow_download_status": {
          "allow_download": true
        },
        "display_url": "pic.twitter.com/0J6bdNicix",
        "expanded_url": "https://twitter.com/gkisokay/status/2046935822096855379/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": []
          },
          "orig": {
            "faces": []
          }
        },
        "id_str": "2046931750341980160",
        "indices": [
          280,
          303
        ],
        "media_key": "3_2046931750341980160",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARxoKZTymjAACgACHGgtSPoa8VMAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHGgplPKaMAAKAAIcaC1I+hrxUwAA",
            "media_key": "3_2046931750341980160"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HGgplPKaMAAuWFr.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 573,
              "w": 1024,
              "x": 0,
              "y": 0
            },
            {
              "h": 1024,
              "w": 1024,
              "x": 0,
              "y": 0
            },
            {
              "h": 1167,
              "w": 1024,
              "x": 0,
              "y": 0
            },
            {
              "h": 1536,
              "w": 768,
              "x": 0,
              "y": 0
            },
            {
              "h": 1536,
              "w": 1024,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1536,
          "width": 1024
        },
        "sizes": {
          "large": {
            "h": 1536,
            "w": 1024
          }
        },
        "type": "photo",
        "url": "https://t.co/0J6bdNicix"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [],
    "user_mentions": []
  },
  "quoted_tweet": {
    "type": "tweet",
    "id": "2046562542202536367",
    "url": "https://x.com/gkisokay/status/2046562542202536367",
    "twitterUrl": "https://twitter.com/gkisokay/status/2046562542202536367",
    "text": "The Local LLM cheat sheet for your 16GB RAM device \n\nI pulled together a lineup of small models that can run comfortably on a Mac Mini or personal laptop while still leaving room for context without melting your machine.\n\nModels for Daily Use \n\nQwen3.5 9B / GGUF / Q4_K_M\nDaily driver. General chat, drafting, research, translation. If you're keeping only one, keep this.  \n\nDeepSeek-R1 Distill Qwen 7B / GGUF / Q4_K_M\nReasoning engine. Math, logic, step-by-step problems. Slower, but worth it when you need actual thinking.\n\nModels for Specialty Work \n\nQwen2.5 Coder 7B / GGUF / Q4_K_M\nCode specialist. Completions, refactors, debugging, repo Q&A. Better than a generalist when the task is code.\n\nLlama 3.1 8B / GGUF / Q4_K_M\nLong context worker. RAG, doc chat, codebase Q and A. The output isn't top tier, but the context is strong for its size. \n\nPhi-4 Mini Reasoning / GGUF / Q4_K_M\nCompact thinker. Logic, structured answers, math, and short coding bursts. Smaller context is the catch.  \n\nModels for Efficiency \n\nGemma 4 E4B / GGUF / Q4_K_M\nLight all-rounder. Writing, chat, light agents, structured output.  \n\nPhi-3.5 Mini / GGUF / Q5_K_M\nPocket sidekick. Summaries, extraction, background doc chat. Easy to pair with a bigger model.  \n\nQwen3.5 2B / GGUF / Q4_K_M\nUseful for summaries, tagging, rewrites, and lightweight sidekick work.  \n\nMicro Models\n\nQwen3.5 0.8B / GGUF / Q5_K_M\nClassification, keyword routing, binary decisions, triage. \n\nGemma 4 E2B-it / GGUF / Q4_K_M\nLightweight chat, quick Q and A, summaries, tiny agents.  \n\nMy personal choice for a single model is Qwen3.5 9B  \n\nFor two models use Qwen3.5 9B + Qwen2.5 Coder 7B for code, or Qwen3.5 9B + Phi-3.5 Mini for support tasks.  \n\nLet me know in the comments your experience with these models, or any I have left out.",
    "source": "Twitter for iPhone",
    "retweetCount": 346,
    "replyCount": 98,
    "likeCount": 2303,
    "quoteCount": 21,
    "viewCount": 413191,
    "createdAt": "Tue Apr 21 12:11:48 +0000 2026",
    "lang": "en",
    "bookmarkCount": 3785,
    "isReply": false,
    "inReplyToId": null,
    "conversationId": "2046562542202536367",
    "displayTextRange": [
      0,
      277
    ],
    "inReplyToUserId": null,
    "inReplyToUsername": null,
    "author": {
      "type": "user",
      "userName": "gkisokay",
      "url": "https://x.com/gkisokay",
      "twitterUrl": "https://twitter.com/gkisokay",
      "id": "629466946",
      "name": "Graeme",
      "isVerified": false,
      "isBlueVerified": true,
      "verifiedType": null,
      "profilePicture": "https://pbs.twimg.com/profile_images/1988420071824470016/nrXOtFnM_normal.jpg",
      "coverPicture": "https://pbs.twimg.com/profile_banners/629466946/1762782644",
      "description": "",
      "location": "",
      "followers": 25399,
      "following": 3681,
      "status": "",
      "canDm": true,
      "canMediaTag": true,
      "createdAt": "Sat Jul 07 16:36:04 +0000 2012",
      "entities": {
        "description": {
          "urls": []
        },
        "url": {}
      },
      "fastFollowersCount": 0,
      "favouritesCount": 26300,
      "hasCustomTimelines": true,
      "isTranslator": false,
      "mediaCount": 2275,
      "statusesCount": 22451,
      "withheldInCountries": [],
      "affiliatesHighlightedLabel": {},
      "possiblySensitive": false,
      "pinnedTweetIds": [
        "2037902655016804496"
      ],
      "profile_bio": {
        "description": "AI agent enjoyer | Founder @amplifi_now | Helping businesses scale with AI",
        "entities": {
          "description": {
            "user_mentions": [
              {
                "id_str": "",
                "indices": [
                  27,
                  39
                ],
                "name": "",
                "screen_name": "amplifi_now"
              }
            ]
          },
          "url": {
            "urls": [
              {
                "display_url": "gkisokay.com",
                "expanded_url": "http://gkisokay.com",
                "indices": [
                  0,
                  23
                ],
                "url": "https://t.co/VGu1PsM6b0"
              }
            ]
          }
        }
      },
      "isAutomated": false,
      "automatedBy": null
    },
    "extendedEntities": {
      "media": [
        {
          "allow_download_status": {
            "allow_download": true
          },
          "display_url": "pic.twitter.com/R6yh87c96V",
          "expanded_url": "https://twitter.com/gkisokay/status/2046562542202536367/photo/1",
          "ext_media_availability": {
            "status": "Available"
          },
          "features": {
            "large": {
              "faces": []
            },
            "orig": {
              "faces": []
            }
          },
          "id_str": "2046562529200185344",
          "indices": [
            278,
            301
          ],
          "media_key": "3_2046562529200185344",
          "media_results": {
            "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARxm2cbzmiAACgACHGbZyfqaQa8AAA==",
            "result": {
              "__typename": "ApiMedia",
              "id": "QXBpTWVkaWE6DAABCgABHGbZxvOaIAAKAAIcZtnJ+ppBrwAA",
              "media_key": "3_2046562529200185344"
            }
          },
          "media_url_https": "https://pbs.twimg.com/media/HGbZxvOaIAA0lHL.jpg",
          "original_info": {
            "focus_rects": [
              {
                "h": 1222,
                "w": 2182,
                "x": 0,
                "y": 0
              },
              {
                "h": 2182,
                "w": 2182,
                "x": 0,
                "y": 0
              },
              {
                "h": 2487,
                "w": 2182,
                "x": 0,
                "y": 0
              },
              {
                "h": 4096,
                "w": 2048,
                "x": 0,
                "y": 0
              },
              {
                "h": 4096,
                "w": 2182,
                "x": 0,
                "y": 0
              }
            ],
            "height": 4096,
            "width": 2182
          },
          "sizes": {
            "large": {
              "h": 2048,
              "w": 1091
            }
          },
          "type": "photo",
          "url": "https://t.co/R6yh87c96V"
        }
      ]
    },
    "card": null,
    "place": {},
    "entities": {
      "hashtags": [],
      "symbols": [],
      "urls": [],
      "user_mentions": []
    },
    "quoted_tweet": null,
    "retweeted_tweet": null,
    "isLimitedReply": false,
    "communityInfo": null,
    "article": null
  },
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "communityInfo": null,
  "article": null
}