Kilo Code Now Supports the Full GPT-5.1 Model Family
Here's how to get started.
OpenAI has just released 4 new models: GPT-5.1, GPT-5.1 Chat, GPT-5.1-Codex and GPT-5.1-Codex-Mini.
All four models have a 400k context window, with the exception of GPT-5.1 Chat (200k context window).
All 4 models are now available in Kilo Code. You can use them either through our Kilo Code gateway (your first $10 top-up gets you +$20 extra) or directly through OpenAI’s API.
Just type “GPT-5.1” when selecting your model inside Kilo Code and you’ll see this list:
Let’s take a deeper look at the models.
GPT-5.1 - What’s new
OpenAI has optimized GPT-5.1 to spend as few tokens as possible.
According to OpenAI, “on straightforward tasks, GPT‑5.1 spends fewer tokens thinking, enabling snappier product experiences and lower token bills. On difficult tasks that require extra thinking, GPT‑5.1 remains persistent, exploring options and checking its work in order to maximize reliability.”
Here’s an example:
GPT-5.1 has implications for coding, “reflected in significant improvements on math and coding evaluations like AIME 2025 and Codeforces.”
Other changes include:
“No reasoning” mode: The reasoning_effort = none option delivers GPT-5.1 intelligence at GPT-4.1 speeds.
Extended prompt caching: Active prompts persist for 24 hours, minimizing both costs and response times in lengthy interactions.
GPT-5.1-Codex & GPT-5.1-Codex Mini
GPT-5.1-Codex is a small incremental update to GPT-5-Codex.
Use GPT-5.1-Mini as backup: According to OpenAI, GPT-5-Codex-Mini allows roughly 4x more usage than GPT-5-Codex, at a slight capability tradeoff due to the more compact model. Here’s the benchmark for GPT-5-Codex:
OpenAI says to “Select GPT-5-Codex-Mini for easier tasks or to extend usage when you’re close to hitting rate limits.”
Kilo Code supports different modes and people are successfully combining GPT-5.1 with GPT-5.1. Codex, using GPT-5.1 for planning and GPT-5.1-Codex for execution.
No pricing changes
The 5.1 family of models are available to all paid API tiers with the same pricing and rate limits as GPT-5.
Here’s the full pricing table:
You can use these models inside our extensions (for VS Code and JetBrains) and the CLI.