repoGitHubTrust 82 · PrimaryPublished 6d agoLive · 2d ago
jaeseok614/llm-gpu-checker-ko
AI hardware fit calculator for LLM, embedding, reranker, OCR and VLM workloads — VRAM, throughput, licensing and multi-GPU planning
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 56%Kicking off GPU Mode [D] →
- PossiblePossibly related (embedding) · 56%Going from single GPU to dual GPU is nice but not in the way I expected →
- PossiblePossibly related (embedding) · 56%I mapped which local LLMs actually fit each RAM tier, 8 to 128GB (open dataset) →
- PossiblePossibly related (embedding) · 55%Best Local LLMs You Can Run on a Single 24GB GPU in 2026: Qwen, Gemma, Mistral, DeepSeek Compared - MarkTechPost →
- PossiblePossibly related (embedding) · 50%Qwen 3.8 27B at 50 tok/s with 100k Context on a 16GB GPU! (beellama.cpp) →
- PossiblePossibly related (embedding) · 51%Single GPU Achieves 15000x Efficiency Boost: Claude Completes Self-Alignment via 48 Hours Non-Stop Intensive Training - 36 Kr →
Covers
newsKicking off GPU Mode [D]newsGoing from single GPU to dual GPU is nice but not in the way I expectednewsI mapped which local LLMs actually fit each RAM tier, 8 to 128GB (open dataset)newsBest Local LLMs You Can Run on a Single 24GB GPU in 2026: Qwen, Gemma, Mistral, DeepSeek Compared - MarkTechPost
Covers (incoming)
Related across the graph
newsI mapped which local LLMs actually fit each RAM tier, 8 to 128GB (open dataset)newsQwen 3.8 27B at 50 tok/s with 100k Context on a 16GB GPU! (beellama.cpp)newsKicking off GPU Mode [D]newsBest Local LLMs You Can Run on a Single 24GB GPU in 2026: Qwen, Gemma, Mistral, DeepSeek Compared - MarkTechPostnewsSingle GPU Achieves 15000x Efficiency Boost: Claude Completes Self-Alignment via 48 Hours Non-Stop Intensive Training - 36 KrnewsGoing from single GPU to dual GPU is nice but not in the way I expected