Check a rig

Choose a listed machine, or type a VRAM or unified-memory size. The query string is the share link: ?device=rtx-4090 or ?memory=24&pool=vram.

Estimated memory is parameter count × bytes per weight × 1.20 when no GGUF file size is stored. Q4 uses 0.5 byte per weight, Q8 uses 1, and FP16 uses 2. The 20% is a modest allowance for KV cache and runtime overhead, not a budget for a 128K or million-token context. When a matching GGUF file is stored, the fit check uses that file size plus an FP16 KV-cache estimate at 8,192 tokens.