Can I run LFM2.5-24B-A2B-Turbo-Brilliance-50X-NEO-MAX-Imatrix-GGUF on an RTX 3060 12 GB?
Yes, just
Its IQ3_M file (10.5 GB) fits in the 10.8 GB an RTX 3060 12 GB can use, with room for ≈ 13K tokens of context.
Full analysis of LFM2.5-24B-A2B-Turbo-Brilliance-50X-NEO-MAX-Imatrix-GGUF
Everything that runs on RTX 3060 12 GB
Get the IQ3_M file on Hugging Face
ollama run hf.co/DavidAU/LFM2.5-24B-A2B-Turbo-Brilliance-50X-NEO-MAX-Imatrix-GGUF:IQ3_MFiles and context on RTX 3060 12 GB
Each file of LFM2.5-24B-A2B-Turbo-Brilliance-50X-NEO-MAX-Imatrix-GGUF (and of GGUF versions the site has analyzed), and how much context fits next to it, with the context cache in 16-bit or 8-bit.
| File | Size | On RTX 3060 12 GB | Context (16-bit) | Context (8-bit) |
|---|---|---|---|---|
| Q8_0 | 25.7 GB | Doesn't fit | — | — |
| Q6_K | 20.0 GB | Doesn't fit | — | — |
| LFM2.5-24B-A2B-TBrill-NEO-Q6_K | 19.7 GB | Doesn't fit | — | — |
| IQ4_XS | 12.8 GB | Doesn't fit | — | — |
| IQ3_M | 10.5 GB | Tight | 13K | 26K |