Update README.md

README.md

@@ -11,4 +11,22 @@ Impact on performance
 We evaluated the models using a panel of giga-models (GPT-4o, Gemini Pro 1.5, and Claude-Sonnet 3.5). The scoring ranged from 0, indicating a model unsuitable
 for the task, to 5, representing a model that fully met expectations. The evaluation was based on 67 instructions across four programming languages: Python,
 Java, JavaScript, and Pseudo-code. All tests were conducted in a French-language context, and models were heavily penalized if they responded in another language,
-even if the response was technically correct.
+even if the response was technically correct.
+
+| model                                           |   score     |
+|:------------------------------------------------|------------:|
+| gemini-1.5-pro                                  | 4.50995     |
+| gpt-4o                                          | 4.50995     |
+| claude3.5-sonnet                                | 4.48756     |
+| deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct     | 4.23691     |
+| meta-llama/Meta-Llama-3.1-70B-Instruct          | 4.23383     |
+| cmarkea/Meta-Llama-3.1-70B-Instruct-4bit        | 4.1393      |
+| cmarkea/Mixtral-8x7B-Instruct-v0.1-4bit         | 3.801       |
+| meta-llama/Meta-Llama-3.1-8B-Instruct           | 3.72637     |
+| mistralai/Mixtral-8x7B-Instruct-v0.1            | 3.33416     |
+| codellama/CodeLlama-13b-Instruct-hf             | 3.32836     |
+| codellama/CodeLlama-34b-Instruct-hf             | 3.27363     |
+| codellama/CodeLlama-7b-Instruct-hf              | 3.19403     |
+| cmarkea/CodeLlama-34b-Instruct-hf-4bit          | 3.12189     |
+| codellama/CodeLlama-70b-Instruct-hf             | 1.81592     |
+| **cmarkea/CodeLlama-70b-Instruct-hf-4bit**      | **1.6409**  |