Provenance and Attribution
Provenance and Attribution (P+A) inference-time protection for code generation
Last updated
Provenance and Attribution (P+A) inference-time protection for code generation
Last updated
Provenance
Tracing code origins. Tabnine’s Provenance feature detects if generated code mirrors existing code, assessing potential copyright infringements.
Attribution identifies and displays the original source of code, along with its licensing, ensuring the user understands usage rights.
Tabnine will perform provenance tracing for all code generated by the Tabnine Chat. If a code block has origins in any open source code, a Code provenance attribution section will appear below the block detailing the source and corresponding licenses.
Users can view the traced source by clicking any line within the Code provenance attribution section.
Enterprise only Private Preview: Reach out to your Customer Success Manager if you wish to participate.
Available for private installations only.
Tabnine Chat support for Code Completions and Inline Actions will be available in 2025
System requirements: P+A requires up to 10TB of free storage
Supported languages: Python, Java, JavaScript, TypeScript, Rust, C, C++
Code match criteria: At least 150 characters, and multiline
Matching index: GitHub is fully indexed, with periodic updates ensuring coverage extends beyond the model’s training data
