Sign in with ChatGPT

Cognition2 min read

Starting today, you can now sign in to Devin with ChatGPT.

If you have an eligible ChatGPT Plus or Pro plan, you can sign in to Devin with ChatGPT and use that plan for any OpenAI models usage in Devin. It works in Devin Cloud, Devin Desktop, and Devin CLI.

Usage in Devin counts toward the Codex and ChatGPT Work usage already included in your plan. You can let Devin use all of it, or set a lower limit for Devin in your ChatGPT settings. If you reach that limit, your sessions keep running on your Devin usage until it resets.

Devin Cloud

OpenAI models are part of the model mix Devin Cloud uses in Lite, Normal, Ultra, and Fusion modes. Once you connect your ChatGPT account, your plan pays for the OpenAI-model share of each session while other models draw down from your included Devin quota. Devin also takes your connected plan into account when choosing models, leaning more on OpenAI models in the mix so that more of each session runs on the usage you already have.

GPT models are particularly strong at computer use. Devin’s testing agent may use them when it opens your app in a browser and clicks through it to check its own work.

You can also try GPT-6 Sol, available now in Devin Cloud as a research preview.

Devin Desktop and CLI

On your machine, you choose the model yourself. Pick any GPT model in Devin Desktop or Devin CLI, and the requests are billed to your ChatGPT plan.

To make your plan go further, use Fusion with GPT-6 Astra as the lead and SWE-2 as the sidekick. The lead plans and reviews, and hands most of the implementation work to the free sidekick.

As a result, Astra spends fewer tokens per task and your ChatGPT plan will last longer. The Artificial Analysis Coding Agent Index shows that using Fusion with Astra and SWE-2 is 39% less expensive than running a session with Astra alone.

Artificial Analysis Coding Agent Index

v1.5
Scorehigher is better
0204060Claude Code, Fable 5.1 (max): 62.262.2Claude CodeFable 5.1 (max)Devin Fusion, Fable 5.1 + SWE-2: 61.761.7+Devin FusionFable 5.1 + SWE-2Codex, Astra (max): 61.661.6CodexAstra (max)Claude Code, Opus 5 (max): 6060Claude CodeOpus 5 (max)Devin Fusion, Astra + SWE-2: 58.958.9+Devin FusionAstra + SWE-2Muse Code, Muse Spark 1.3 (max): 5454Muse CodeMuse Spark 1.3 (max)Opencode, GLM-5.3: 5454OpencodeGLM-5.3Kimi Code CLI, Kimi K3: 5252Kimi Code CLIKimi K3Grok Build, Grok 4.6 (xhigh): 4747Grok BuildGrok 4.6 (xhigh)Claude Code, Qwen3.8 Max: 4343Claude CodeQwen3.8 MaxCodex, DeepSeek V4: 4343CodexDeepSeek V4Antigravity SDK, Gemini 3.8 Flash: 4242Antigravity SDKGemini 3.8 Flash
Costlower is better
Claude Code: 100 ($12.36); Devin Fusion: 64 ($7.90); 36% lower costClaude CodeFable 5.1−36%+Devin FusionFable 5.1 + SWE-2Codex: 100 ($7.47); Devin Fusion: 61 ($4.54); 39% lower costCodexAstra−39%+Devin FusionAstra + SWE-2

Get started

Sign in to Devin with ChatGPT today, and try OpenAI models in Devin Cloud, Devin Desktop, or Devin CLI.

Available on Devin Pro, Max, and Teams plans. For setup steps, eligible plans, and how billing works across modes, see the docs.