mradermacher commited on
Commit
d351223
·
verified ·
1 Parent(s): 43c2645

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -0
README.md CHANGED
@@ -38,7 +38,10 @@ more details, including on how to concatenate multi-part files.
38
  | [P1](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part1of7) [P2](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part2of7) [P3](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part3of7) [P4](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part4of7) [P5](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part5of7) [P6](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part6of7) [P7](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part7of7) | i1-IQ2_S | 315.4 | |
39
  | [P1](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part1of7) [P2](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part2of7) [P3](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part3of7) [P4](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part4of7) [P5](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part5of7) [P6](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part6of7) [P7](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part7of7) | i1-IQ2_M | 343.0 | |
40
  | [P1](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part1of8) [P2](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part2of8) [P3](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part3of8) [P4](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part4of8) [P5](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part5of8) [P6](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part6of8) [P7](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part7of8) [P8](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part8of8) | i1-Q2_K | 374.5 | IQ3_XXS probably better |
 
 
41
  | [P1](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part01of12) [P2](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part02of12) [P3](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part03of12) [P4](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part04of12) [P5](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part05of12) [P6](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part06of12) [P7](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part07of12) [P8](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part08of12) [P9](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part09of12) [P10](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part10of12) [P11](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part11of12) [P12](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part12of12) | i1-Q4_K_S | 578.1 | optimal size/speed/quality |
 
42
 
43
  Here is a handy graph by ikawrakow comparing some lower-quality quant
44
  types (lower is better):
 
38
  | [P1](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part1of7) [P2](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part2of7) [P3](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part3of7) [P4](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part4of7) [P5](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part5of7) [P6](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part6of7) [P7](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_S.gguf.part7of7) | i1-IQ2_S | 315.4 | |
39
  | [P1](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part1of7) [P2](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part2of7) [P3](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part3of7) [P4](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part4of7) [P5](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part5of7) [P6](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part6of7) [P7](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ2_M.gguf.part7of7) | i1-IQ2_M | 343.0 | |
40
  | [P1](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part1of8) [P2](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part2of8) [P3](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part3of8) [P4](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part4of8) [P5](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part5of8) [P6](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part6of8) [P7](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part7of8) [P8](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q2_K.gguf.part8of8) | i1-Q2_K | 374.5 | IQ3_XXS probably better |
41
+ | [P1](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ3_XS.gguf.part01of10) [P2](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ3_XS.gguf.part02of10) [P3](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ3_XS.gguf.part03of10) [P4](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ3_XS.gguf.part04of10) [P5](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ3_XS.gguf.part05of10) [P6](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ3_XS.gguf.part06of10) [P7](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ3_XS.gguf.part07of10) [P8](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ3_XS.gguf.part08of10) [P9](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ3_XS.gguf.part09of10) [P10](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-IQ3_XS.gguf.part10of10) | i1-IQ3_XS | 416.4 | |
42
+ | [P1](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q3_K_M.gguf.part01of10) [P2](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q3_K_M.gguf.part02of10) [P3](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q3_K_M.gguf.part03of10) [P4](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q3_K_M.gguf.part04of10) [P5](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q3_K_M.gguf.part05of10) [P6](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q3_K_M.gguf.part06of10) [P7](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q3_K_M.gguf.part07of10) [P8](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q3_K_M.gguf.part08of10) [P9](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q3_K_M.gguf.part09of10) [P10](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q3_K_M.gguf.part10of10) | i1-Q3_K_M | 490.0 | IQ3_S probably better |
43
  | [P1](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part01of12) [P2](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part02of12) [P3](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part03of12) [P4](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part04of12) [P5](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part05of12) [P6](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part06of12) [P7](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part07of12) [P8](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part08of12) [P9](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part09of12) [P10](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part10of12) [P11](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part11of12) [P12](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_S.gguf.part12of12) | i1-Q4_K_S | 578.1 | optimal size/speed/quality |
44
+ | [P1](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_M.gguf.part01of13) [P2](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_M.gguf.part02of13) [P3](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_M.gguf.part03of13) [P4](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_M.gguf.part04of13) [P5](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_M.gguf.part05of13) [P6](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_M.gguf.part06of13) [P7](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_M.gguf.part07of13) [P8](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_M.gguf.part08of13) [P9](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_M.gguf.part09of13) [P10](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_M.gguf.part10of13) [P11](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_M.gguf.part11of13) [P12](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_M.gguf.part12of13) [P13](https://huggingface.co/mradermacher/BigLlama-3.1-1T-Instruct-i1-GGUF/resolve/main/BigLlama-3.1-1T-Instruct.i1-Q4_K_M.gguf.part13of13) | i1-Q4_K_M | 610.5 | fast, recommended |
45
 
46
  Here is a handy graph by ikawrakow comparing some lower-quality quant
47
  types (lower is better):