Update README.md
Browse files
README.md
CHANGED
|
@@ -13,20 +13,20 @@ for the task, to 5, representing a model that fully met expectations. The evalua
|
|
| 13 |
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,
|
| 14 |
even if the response was technically correct.
|
| 15 |
|
| 16 |
-
| model | score |
|
| 17 |
-
|
| 18 |
-
| gemini-1.5-pro | 4.50995 |
|
| 19 |
-
| gpt-4o | 4.50995 |
|
| 20 |
-
| claude3.5-sonnet | 4.48756 |
|
| 21 |
-
| deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct | 4.23691 |
|
| 22 |
-
| meta-llama/Meta-Llama-3.1-70B-Instruct | 4.23383 |
|
| 23 |
-
| cmarkea/Meta-Llama-3.1-70B-Instruct-4bit | 4.1393 |
|
| 24 |
-
| cmarkea/Mixtral-8x7B-Instruct-v0.1-4bit | 3.801 |
|
| 25 |
-
| meta-llama/Meta-Llama-3.1-8B-Instruct | 3.72637 |
|
| 26 |
-
| mistralai/Mixtral-8x7B-Instruct-v0.1 | 3.33416 |
|
| 27 |
-
| codellama/CodeLlama-13b-Instruct-hf | 3.32836 |
|
| 28 |
-
| codellama/CodeLlama-34b-Instruct-hf | 3.27363 |
|
| 29 |
-
| codellama/CodeLlama-7b-Instruct-hf | 3.19403 |
|
| 30 |
-
| cmarkea/CodeLlama-34b-Instruct-hf-4bit | 3.12189 |
|
| 31 |
-
| codellama/CodeLlama-70b-Instruct-hf | 1.81592 |
|
| 32 |
-
| **cmarkea/CodeLlama-70b-Instruct-hf-4bit** | **1.6409** |
|
|
|
|
| 13 |
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,
|
| 14 |
even if the response was technically correct.
|
| 15 |
|
| 16 |
+
| model | score | # params (Billion) | size (GB) |
|
| 17 |
+
|------------------------------------------------:|:-----------:|:------------------:|:---------:|
|
| 18 |
+
| gemini-1.5-pro | 4.50995 | NA | NA |
|
| 19 |
+
| gpt-4o | 4.50995 | NA | NA |
|
| 20 |
+
| claude3.5-sonnet | 4.48756 | NA | NA |
|
| 21 |
+
| deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct | 4.23691 | 15.7 | 31.4 |
|
| 22 |
+
| meta-llama/Meta-Llama-3.1-70B-Instruct | 4.23383 | 70.06 | 140.12 |
|
| 23 |
+
| cmarkea/Meta-Llama-3.1-70B-Instruct-4bit | 4.1393 | 70.06 | 35.3 |
|
| 24 |
+
| cmarkea/Mixtral-8x7B-Instruct-v0.1-4bit | 3.801 | 46.7 | 23.35 |
|
| 25 |
+
| meta-llama/Meta-Llama-3.1-8B-Instruct | 3.72637 | 8.03 | 16.06 |
|
| 26 |
+
| mistralai/Mixtral-8x7B-Instruct-v0.1 | 3.33416 | 46.7 | 93.4 |
|
| 27 |
+
| codellama/CodeLlama-13b-Instruct-hf | 3.32836 | 13 | 26 |
|
| 28 |
+
| codellama/CodeLlama-34b-Instruct-hf | 3.27363 | 33.7 | 67.4 |
|
| 29 |
+
| codellama/CodeLlama-7b-Instruct-hf | 3.19403 | 6.74 | 13.48 |
|
| 30 |
+
| cmarkea/CodeLlama-34b-Instruct-hf-4bit | 3.12189 | 33.7 | 16.85 |
|
| 31 |
+
| codellama/CodeLlama-70b-Instruct-hf | 1.81592 | 69 | 138 |
|
| 32 |
+
| **cmarkea/CodeLlama-70b-Instruct-hf-4bit** | **1.6409** | 69 | 34.5 |
|