Qodo Unveils Efficient Code Embedding Model

Qodo has announced the release of Qodo-Embed-1-1.5B, a new open-source code embedding model that sets a new standard for efficiency and performance, announced in a press release. This model, with 1.5 billion parameters, surpasses larger models from OpenAI and Salesforce in industry benchmarks.

The Qodo-Embed-1-1.5B model is designed to enhance code search, retrieval, and understanding, making it particularly beneficial for enterprise development teams managing extensive codebases. It achieves superior results on the Code Information Retrieval Benchmark (CoIR), scoring 70.06, compared to Salesforce's 67.41 and OpenAI's 65.17.

Qodo's model is available on Hugging Face under the OpenRAIL++-M license, allowing developers to integrate it into their workflows. For enterprises requiring additional capabilities, larger versions are available under commercial licensing. The model will also be accessible through NVIDIA's NIM platform and AWS SageMaker JumpStart, facilitating easy deployment and integration into existing development environments.