the model store, reimagined
Two panels. One question: does it fit my machine?
Browse official models with real creator logos on the left, read their README on the right. Every download option tells you whether it fits your RAM and VRAM — before you download.
llama-3.1-8b-instruct.Q4_K_M.gguf
llama-3.1-8b-instruct.Q5_K_S.gguf
llama-3.1-8b-instruct.fp16.gguf
→ balanced · recommended · ✓ fits your GPU
01
Two-panel layout
Model list with real company logos on the left, rich detail view with README on the right. No more scrolling through walls of text.
02
Hardware-fit badge
"Fits your device — fast on GPU", "CPU OK", or "Too large" — computed from your detected RAM and VRAM. No surprises after a 30-minute download.
03
Plain-language quants
"Balanced quality", "High quality", or "Smallest size" instead of Q4_K_M.gguf. Sorted by what fits your machine best.
04
Filters & tags
Search by size (1–8B, 8–14B, 14B+) or capability (Tools, Vision, Code). Every model shows what it can do with clear icons.