CodeRabbit uses GPT-6.1 Sol to assess security risks in codebases

News

OpenAI says CodeRabbit is using GPT-6.1 Sol to review security risks and proposed fixes across complex codebases.

CodeRabbit is using OpenAI’s GPT-6.1 Sol to investigate security risks across complex codebases, according to OpenAI in a post published on Monday, 26 October 2026. OpenAI said the security agent helps developers review findings and proposed fixes, underscoring a use case focused on code analysis rather than general chat.

Evidence and context

The announcement builds on OpenAI’s earlier introduction of GPT-6.1 Sol on 29 September 2026. OpenAI’s own materials say the model was designed for higher-capability reasoning and coding tasks, and list a 1,050,000-token context window, 128,000 maximum output tokens, text input and output, and image input in current API documentation. Standard pricing listed in the documentation is $2 per million input tokens, $0.10 per million cached input tokens and $10 per million output tokens, with additional pricing conditions applying.

The enrichment provided alongside the announcement also notes that OpenAI’s reported evaluations versus GPT-6 Sol showed gains on several benchmarks, including DeepSWE v1.1, AutomationBench and OSWorld 2.0, as well as lower error rates on deliberately difficult factuality prompts. Those figures are vendor-reported rather than independently verified reproductions, and they are not the same as CodeRabbit’s own deployment results.

Market data supplied with the source refers to Solana, not the OpenAI model. As of 26 October 2026, SOL traded at $100.61, down 0.50% on the day, with a market value of $59.36 billion and no 24-hour volume figure provided. The materials supplied do not establish any official cryptocurrency connection between SOL and GPT-6.1 Sol.

OpenAI also said on 21 October 2026 that its models are coming to Google Cloud, a separate development that may broaden access to the company’s model lineup. The evidence provided does not include terms of CodeRabbit’s deployment, customer impact metrics, or a verified timetable for further rollout.