{"slug": "ai-hardware-fit-check-whether-a-local-model-fits-then-compare-gpus", "title": "AI Hardware Fit: check whether a local model fits, then compare GPUs", "summary": "The AI Hardware Fit calculator received a November update that expands its catalog to 160 hardware presets and 350 models, adding MiniCPM5 2B, LFM2.5 24B A2B and LFM3 1.7B along with full-system budget estimates using personal-use defaults. The tool now carries an MLX-VLM 0.7.11+ recipe for the official LFM3 1.7B Apple Silicon checkpoint, using verified 4-bit weights and showing a verification notice for other precisions rather than silently switching to 4-bit. Developer jaeseok614 is soliciting reproducible benchmark reports covering hardware, quantization, runtime version, context length, peak memory and separate prompt-processing and decode speeds, with results that disagree with the estimates described as especially helpful.", "body_md": "A November update for anyone planning a local setup: the catalog now has 160 hardware presets and 350 models.\n\nRecent additions include MiniCPM5 2B, LFM2.5 24B A2B, and LFM3 1.7B. I have also added full-system budget estimates with personal-use defaults, so you can compare the GPU alongside the supporting components.\n\nOne concrete compatibility update today: the official LFM3 1.7B Apple Silicon checkpoint now has an MLX-VLM 0.7.11+ recipe. The calculator uses that runtime and the verified 4-bit weights. Other precisions show a verification notice instead of silently switching to 4-bit.\n\nThe two frontier charts are still there: VRAM versus published benchmark quality, and GPU reference price versus estimated speed. Memory and speed estimates include their assumptions, and publisher evaluations have source links.\n\nIf you try it on a Mac mini, Mac Studio, or a consumer GPU, I would appreciate a reproducible result: hardware and memory, model/checkpoint, quantization, runtime/version, context length, peak memory, and separate prompt-processing and decode speeds. A result that disagrees with the estimate is especially helpful.\n\nDemo: [AI Hardware Fit — 내 GPU에서 돌아가는 AI 모델 찾기](https://jaeseok614.github.io/ai-hardware-fit/?lang=en)\n\nSource: [GitHub - jaeseok614/ai-hardware-fit: Can my GPU run this? Pick a GPU to get 3 fitting AI models, quantization, VRAM, speed ranges, and run commands. 한국어·English. · GitHub](https://github.com/jaeseok614/ai-hardware-fit)\n\nMeasurement report: [Sign in to GitHub · GitHub](https://github.com/jaeseok614/ai-hardware-fit/issues/new?template=benchmark-report.yml)\n\nOfficial LFM3 MLX instructions: [LiquidAI/LFM3-1.7B-MLX-4bit · Hugging Face](https://huggingface.co/LiquidAI/LFM3-1.7B-MLX-4bit)", "url": "https://wpnews.pro/news/ai-hardware-fit-check-whether-a-local-model-fits-then-compare-gpus", "canonical_source": "https://discuss.huggingface.co/t/ai-hardware-fit-check-whether-a-local-model-fits-then-compare-gpus/182167#post_2", "published_at": "2026-11-10 03:49:20+00:00", "updated_at": "2026-11-10 03:49:45.630218+00:00", "lang": "en", "topics": ["ai-tools", "ai-infrastructure", "large-language-models", "ai-products"], "entities": ["AI Hardware Fit", "jaeseok614", "MiniCPM5 2B", "LFM2.5 24B A2B", "LFM3 1.7B", "MLX-VLM 0.7.11", "LiquidAI", "GitHub"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/ai-hardware-fit-check-whether-a-local-model-fits-then-compare-gpus", "markdown": "https://wpnews.pro/news/ai-hardware-fit-check-whether-a-local-model-fits-then-compare-gpus.md", "text": "https://wpnews.pro/news/ai-hardware-fit-check-whether-a-local-model-fits-then-compare-gpus.txt", "jsonld": "https://wpnews.pro/news/ai-hardware-fit-check-whether-a-local-model-fits-then-compare-gpus.jsonld"}}