We’re excited to roll out support for OpenAI’s latest model, GPT-5.1, inside Warp. After weeks of testing, benchmarking, and tuning, GPT-5.1 has proven so capable that we’ve made it the default model for all new users starting today.
Under the hood of GPT-5.1
GPT-5.1 builds on what we liked about GPT-5. It's very smart and efficient, but notably GPT-5.1 is much faster due to adaptive reasoning. Under the hood, GPT-5.1 adjusts its reasoning automatically based on the complexity of your task. You still always get a smart answer, but you also usually get a much faster answer.
Here’s what stood out most in testing:
- Latency: GPT-5.1 consistently delivers faster results— 15% faster time to first token, and 40% faster task completion (measured on a subset of SWE-bench Verified)
- Quality: Maintains the high bar set by GPT-5
- Deep debugging: GPT-5.1 excels at root-causing and fixing tough, multi-layered problems. The kind that require tracing issues across files, dependencies, and environments.
Zach shows how GPT-5.1 performs on a real coding task.
Working with OpenAI
We worked closely with the OpenAI team to make sure GPT-5.1 performs well with Warp. This included optimizing prompt structure, tool definitions, and the balance between planning and execution. The collaboration helped us achieve a combination of speed, accuracy, and context awareness that feels natural inside the terminal.
Warp's collaboration and feedback has been invaluable in shaping GPT-5.1 to excel even more on coding. With significant gains in code quality, steerability, and speed, GPT-5.1 is performing exceptionally well inside Warp, and we’re thrilled to see it becoming the default experience for new users.
Marc Manara, Head of Startups, OpenAI
Try 5.1 in Warp
To give developers the chance to experience GPT-5.1 firsthand, we’re offering a limited-time promotion. Use the code BUILD510 to get your first month of Warp Build for $5.10. The code is valid for the first 510 users.
