MPWARE commited on
Commit
21b6e66
·
verified ·
1 Parent(s): 35566e4

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +13 -1
README.md CHANGED
@@ -3,4 +3,16 @@ license: mit
3
  base_model:
4
  - deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
5
  ---
6
- AWQ 4 bits quantization from DeepSeek-R1-Distill-Qwen-14B commit 123265213609ea67934b1790bbb0203d3c50f54f
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  base_model:
4
  - deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
5
  ---
6
+ AWQ 4 bits quantization from DeepSeek-R1-Distill-Qwen-14B commit 123265213609ea67934b1790bbb0203d3c50f54f
7
+
8
+ ```python
9
+ from awq import AutoAWQForCausalLM
10
+ from transformers import AutoTokenizer
11
+ model_name = "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B"
12
+ commit_hash = "123265213609ea67934b1790bbb0203d3c50f54f"
13
+ # Download the model and tokenizer at the specific commit hash
14
+ model = AutoAWQForCausalLM.from_pretrained(model_name, revision=commit_hash)
15
+ tokenizer = AutoTokenizer.from_pretrained(model_name, revision=commit_hash)
16
+ quant_config = { "zero_point": True, "q_group_size": 128, "w_bit": 4, "version": "GEMM" }
17
+ model.quantize(tokenizer, quant_config=quant_config)
18
+ ```