runmyllm.com

概況

資料每週更新；估算值說明見 資料說明

AI 解讀

由大模型基於落地頁內容整理，推測性的判斷已註明。

搜尋、外鏈與變現

{
  "domain": "runmyllm.com",
  "name": "RunMyLLM",
  "title": "Local LLM VRAM calculator — which LLM can your GPU or Mac run? | RunMyLLM",
  "description": "The local LLM VRAM calculator: pick your GPU or Apple Silicon chip and see which open-weight LLMs fit — with a recommended model per job (coding, reasoning, vision, agents, speed), quantised weight sizes, KV cache at your context length, estimated tokens per second, and the command to run it. Catalogue updated weekly.",
  "desc_zh": "本地大模型顯存計算器，選好 GPU 或 Apple 晶片即可查看能跑哪些開源模型，並給出推薦與執行命令。",
  "cat1": "[\"编程与开发\", \"研究与数据\"]",
  "cat2": "[\"本地模型部署\", \"显存与硬件选型\", \"开源模型对比\"]",
  "site_type": "[\"关键词驱动\"]",
  "keyword_type": "长尾词",
  "model_family": null,
  "model_version": null,
  "status": "正常",
  "registered": "2026-08-21",
  "rank": 401048,
  "visits": 28801,
  "visits3": "[10071, 10071, 28801]",
  "growth": 186.0,
  "dr": null,
  "tech": "[]",
  "analytics": "[]",
  "payments": "[]",
  "pricing": "[]",
  "monetization": "[]",
  "monetization_score": 36.1,
  "keywords": "[\"local llm vram calculator\", \"which llm can my gpu run\", \"llm vram calculator\", \"best local llm for my vram\", \"run llm on apple silicon\"]",
  "outbound": "[\"aiarticlepricer.com\"]",
  "badges": "[]",
  "seo": "{\"words\": 753, \"lang\": \"en\", \"internal\": 53, \"outbound\": 1, \"images\": 0, \"images_no_alt\": 0, \"schema\": []}",
  "insight": "{\"problem\": \"想本地跑開源大模型的人，不清楚自己的顯卡或 Mac 顯存到底能裝下哪個模型、什麼量化、多長上下文。\", \"before\": \"靠論壇帖、Reddit 討論和零散表格手動估算權重與 KV cache，反覆試錯下載模型。\", \"now\": \"選設備再選用途，直接看到適配模型、量化體積、KV cache、預估 tokens/s 和一條運行命令。\", \"vs\": \"多數同類只算權重體積，它把 KV cache、上下文長度、速度預估和推薦模型按任務場景一起給出。\", \"audience\": [\"本地部署愛好者\", \"AI 獨立開發者\", \"選購 GPU 或 Mac 的開發者\"], \"hooks\": [\"按用途直接給推薦模型，不用自己挑\", \"算上 KV cache 和上下文，結果更貼近實際\", \"附一條可直接執行的運行命令\"], \"pay_willingness\": \"低\", \"growth\": \"訪問量約 2.9 萬、增長 186%，處於快速上升的早期階段，屬於工具型流量起量期。\", \"acquisition\": \"推測主要靠「local llm vram calculator」「which llm can my gpu run」這類長尾搜尋，以及模型目錄週更帶來的復訪。\", \"moat\": \"壁壘偏薄，計算邏輯易被複製；優勢在模型目錄週更、任務向推薦和命令生成形成的使用習慣。\", \"money\": \"頁面未檢測到收款渠道與價格線索，當前大概率免費，未見訂閱或付費入口。\"}",
  "is_sample_revenue": 0,
  "is_new": 1,
  "reg_relaxed": 0,
  "updated_at": "2026-10-18T01:31:32+00:00"
}

同類站點

同一分類中估算流量最接近的站點。