auto-patch README.md

README.md

@@ -11,8 +11,9 @@ extra_gated_prompt: |-
   Access to this model requires reading and agreeing to the following agreement [here](https://github.com/Nanbeige/Nanbeige/blob/main/License_Agreement_for_Large_Language_Models_Nanbeige.pdf)
 language:
 - en
+- zh
 library_name: transformers
-license: other
+license: apache-2.0
 quantized_by: mradermacher
 tags:
 - llm
@@ -46,7 +47,7 @@ more details, including on how to concatenate multi-part files.
 | [GGUF](https://huggingface.co/mradermacher/Nanbeige1.5-8B-Chat-GGUF/resolve/main/Nanbeige1.5-8B-Chat.Q3_K_L.gguf) | Q3_K_L | 4.6 |  |
 | [GGUF](https://huggingface.co/mradermacher/Nanbeige1.5-8B-Chat-GGUF/resolve/main/Nanbeige1.5-8B-Chat.IQ4_XS.gguf) | IQ4_XS | 4.6 |  |
 | [GGUF](https://huggingface.co/mradermacher/Nanbeige1.5-8B-Chat-GGUF/resolve/main/Nanbeige1.5-8B-Chat.Q4_0.gguf) | Q4_0 | 4.8 | fast, low quality |
-| [GGUF](https://huggingface.co/mradermacher/Nanbeige1.5-8B-Chat-GGUF/resolve/main/Nanbeige1.5-8B-Chat.IQ4_NL.gguf) | IQ4_NL | 4.8 | slightly worse than Q4_K_S |
+| [GGUF](https://huggingface.co/mradermacher/Nanbeige1.5-8B-Chat-GGUF/resolve/main/Nanbeige1.5-8B-Chat.IQ4_NL.gguf) | IQ4_NL | 4.8 | prefer IQ4_XS |
 | [GGUF](https://huggingface.co/mradermacher/Nanbeige1.5-8B-Chat-GGUF/resolve/main/Nanbeige1.5-8B-Chat.Q4_K_S.gguf) | Q4_K_S | 4.9 | fast, recommended |
 | [GGUF](https://huggingface.co/mradermacher/Nanbeige1.5-8B-Chat-GGUF/resolve/main/Nanbeige1.5-8B-Chat.Q4_K_M.gguf) | Q4_K_M | 5.1 | fast, recommended |
 | [GGUF](https://huggingface.co/mradermacher/Nanbeige1.5-8B-Chat-GGUF/resolve/main/Nanbeige1.5-8B-Chat.Q5_K_S.gguf) | Q5_K_S | 5.8 |  |
@@ -54,7 +55,6 @@ more details, including on how to concatenate multi-part files.
 | [GGUF](https://huggingface.co/mradermacher/Nanbeige1.5-8B-Chat-GGUF/resolve/main/Nanbeige1.5-8B-Chat.Q6_K.gguf) | Q6_K | 6.8 | very good quality |
 | [GGUF](https://huggingface.co/mradermacher/Nanbeige1.5-8B-Chat-GGUF/resolve/main/Nanbeige1.5-8B-Chat.Q8_0.gguf) | Q8_0 | 8.6 | fast, best quality |
 
-
 Here is a handy graph by ikawrakow comparing some lower-quality quant
 types (lower is better):