bartowski commited on
Commit
8d64014
·
verified ·
1 Parent(s): a78d36f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -0
README.md CHANGED
@@ -33,6 +33,8 @@ The reason is, the FFN (feed forward networks) of gpt-oss do not behave nicely w
33
 
34
  The rest of these are provided for your own interest in case you feel like experimenting, but the size savings is basically non-existent so I would not recommend running them, they are provided simply for show:
35
 
 
 
36
  | [gpt-oss-120b-bf16.gguf](https://huggingface.co/bartowski/openai_gpt-oss-120b-GGUF/tree/main/openai_gpt-oss-120b-bf16) | bf16 | 65.37GB | true | Full BF16 weights. |
37
  | [gpt-oss-120b-Q6_K.gguf](https://huggingface.co/bartowski/openai_gpt-oss-120b-GGUF/tree/main/openai_gpt-oss-120b-Q6_K) | Q6_K | 63.28GB | true | Q6_K with all FFN kept at MXFP4_MOE. |
38
  | [gpt-oss-120b-Q4_K_L.gguf](https://huggingface.co/bartowski/openai_gpt-oss-120b-GGUF/tree/main/openai_gpt-oss-120b-Q4_K_L) | Q4_K_L | 63.06GB | true | Uses Q8_0 for embed and output weights. Q4_K_M with all FFN kept at MXFP4_MOE. |
 
33
 
34
  The rest of these are provided for your own interest in case you feel like experimenting, but the size savings is basically non-existent so I would not recommend running them, they are provided simply for show:
35
 
36
+ | Filename | Quant type | File Size | Split | Description |
37
+ | -------- | ---------- | --------- | ----- | ----------- |
38
  | [gpt-oss-120b-bf16.gguf](https://huggingface.co/bartowski/openai_gpt-oss-120b-GGUF/tree/main/openai_gpt-oss-120b-bf16) | bf16 | 65.37GB | true | Full BF16 weights. |
39
  | [gpt-oss-120b-Q6_K.gguf](https://huggingface.co/bartowski/openai_gpt-oss-120b-GGUF/tree/main/openai_gpt-oss-120b-Q6_K) | Q6_K | 63.28GB | true | Q6_K with all FFN kept at MXFP4_MOE. |
40
  | [gpt-oss-120b-Q4_K_L.gguf](https://huggingface.co/bartowski/openai_gpt-oss-120b-GGUF/tree/main/openai_gpt-oss-120b-Q4_K_L) | Q4_K_L | 63.06GB | true | Uses Q8_0 for embed and output weights. Q4_K_M with all FFN kept at MXFP4_MOE. |