#1
granite-4.1-3b-BF16.gguf
- GGUF size
- 6.34 GiB
- Markdown
- granite-4.1-3b-BF16.md 2.66 KiB
- SVG
- granite-4.1-3b-BF16.svg 1.21 KiB
Each runnable GGUF file was prompted in turn with llama.cpp, saved as Markdown, then the SVG block was extracted into a matching SVG file. Cards are ordered by GGUF file size from largest to smallest.
These files are 135-byte Git LFS pointer stubs rather than local model data, so they were not runnable with llama.cpp.
granite-4.1-3b-UD-Q2_K_XL.gguf 135 Bgranite-4.1-3b-UD-Q4_K_XL.gguf 135 Bgranite-4.1-3b-IQ4_NL.gguf 135 B