Update README.md
Browse files
README.md
CHANGED
@@ -122,9 +122,9 @@ To evaluate the model for vulnerability remediation we used the `Static Analysis
|
|
122 |
|
123 |
| Model | HumanEval | HumanEval Fix Python| Static Analysis Eval |
|
124 |
| ----- | ----------| ------------------- | -------------------- |
|
125 |
-
| GPT-4 | 86.6 | 47 | 55.26 |
|
126 |
| patched-coder-34b | 53.57 | 41.34 | 51.32 |
|
127 |
| CodeLlama-34b-Python | 53.29 | 33.14 | 27.63 |
|
|
|
128 |
|
129 |
Based on the results on these benchmarks, patched-coder-34b is the SOTA open code LLM. Other code LLMs (e.g. from WizardCoder and Phind) are trained on
|
130 |
either unknown proprietary datasets or used OpenAI's APIs for training, thus making them unviable for commercial use.
|
|
|
122 |
|
123 |
| Model | HumanEval | HumanEval Fix Python| Static Analysis Eval |
|
124 |
| ----- | ----------| ------------------- | -------------------- |
|
|
|
125 |
| patched-coder-34b | 53.57 | 41.34 | 51.32 |
|
126 |
| CodeLlama-34b-Python | 53.29 | 33.14 | 27.63 |
|
127 |
+
| GPT-4 | 86.6 | 47 | 55.26 |
|
128 |
|
129 |
Based on the results on these benchmarks, patched-coder-34b is the SOTA open code LLM. Other code LLMs (e.g. from WizardCoder and Phind) are trained on
|
130 |
either unknown proprietary datasets or used OpenAI's APIs for training, thus making them unviable for commercial use.
|