sthaps commited on
Commit
f3221e4
·
verified ·
1 Parent(s): 97c2581

Update Model Card

Browse files
Files changed (1) hide show
  1. README.md +30 -0
README.md ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # DeepSeek-R1-ThaiInsurance-COT-Demo1 - GGUF
2
+
3
+ ## About
4
+ This repository contains GGUF weights for [Pavariss/DeepSeek-R1-ThaiInsurance-COT-Demo1](https://huggingface.co/Pavariss/DeepSeek-R1-ThaiInsurance-COT-Demo1).
5
+
6
+ For a convenient overview and download list, visit our [model page](https://huggingface.co/sthaps/DeepSeek-R1-ThaiInsurance-COT-Demo1).
7
+
8
+ ## Usage
9
+ If you are unsure how to use GGUF files, refer to the [llama.cpp documentation](https://github.com/ggerganov/llama.cpp) for more details.
10
+
11
+ ### Llama.cpp CLI
12
+ ```bash
13
+ ./llama-cli -m DeepSeek-R1-ThaiInsurance-COT-Demo1-q4_k_m.gguf -p "Hello!"
14
+ ```
15
+
16
+ ## Provided Quants
17
+
18
+ (sorted by size, not necessarily quality)
19
+
20
+ | Link | Type | Size/GB | Notes |
21
+ | :--- | :--- | :---: | :--- |
22
+ | [GGUF](https://huggingface.co/sthaps/DeepSeek-R1-ThaiInsurance-COT-Demo1/blob/main/DeepSeek-R1-ThaiInsurance-COT-Demo1-q2_k.gguf) | q2_k | 2.96 | very low quality, for testing |
23
+ | [GGUF](https://huggingface.co/sthaps/DeepSeek-R1-ThaiInsurance-COT-Demo1/blob/main/DeepSeek-R1-ThaiInsurance-COT-Demo1-q3_k_m.gguf) | q3_k_m | 3.74 | |
24
+ | [GGUF](https://huggingface.co/sthaps/DeepSeek-R1-ThaiInsurance-COT-Demo1/blob/main/DeepSeek-R1-ThaiInsurance-COT-Demo1-q4_0.gguf) | q4_0 | 4.34 | |
25
+ | [GGUF](https://huggingface.co/sthaps/DeepSeek-R1-ThaiInsurance-COT-Demo1/blob/main/DeepSeek-R1-ThaiInsurance-COT-Demo1-q4_k_m.gguf) | q4_k_m | 4.58 | recommended, good balance |
26
+ | [GGUF](https://huggingface.co/sthaps/DeepSeek-R1-ThaiInsurance-COT-Demo1/blob/main/DeepSeek-R1-ThaiInsurance-COT-Demo1-q5_k_m.gguf) | q5_k_m | 5.34 | |
27
+ | [GGUF](https://huggingface.co/sthaps/DeepSeek-R1-ThaiInsurance-COT-Demo1/blob/main/DeepSeek-R1-ThaiInsurance-COT-Demo1-q8_0.gguf) | q8_0 | 7.95 | near-full precision |
28
+
29
+ ## Thanks
30
+ Special thanks to the `llama.cpp` team for their amazing work.