On Oct 9 Cognition said Devin can now link any personal ChatGPT plan (Go, Plus or Pro). The GPT share of each session is billed to the ChatGPT subscription instead of Devin quota or on-demand credits, across Devin Cloud, Desktop and CLI. It requires a self-serve Devin Pro, Max or Teams plan (not Free or Enterprise), and falls back to Devin quota when the ChatGPT allowance runs out.

Key Takeaways

  • ✓Eligible ChatGPT plans widened from Plus/Pro at the Sep 29 launch to Go, Plus and Pro; free ChatGPT accounts can't share usage
  • ✓Only self-serve Devin Pro ($20/mo), Max ($200/mo) and Teams ($80/mo minimum); not Free or Enterprise
  • ✓GPT usage bills only to ChatGPT, never double-counted against Devin quota; fast/priority variants excluded
  • ✓Vendor-cited Artificial Analysis Coding Agent Index v1.5: Fusion (GPT-6 Astra + SWE-2) scores 58.9 at $4.54/task vs Codex Astra (max) 61.6 at $7.47, 39% cheaper
Devin now works with any personal ChatGPT plan: Go, Plus or Pro subscriptions cover GPT model usage instead of Devin quota
🖼️Official Media
Click to view high-res
🧭

Turn your technical choice into a development budget

Compare 40 dev plans & simulate token costs vs $20/mo subscriptions

🔬

In-Depth Technical Analysis

Cognition now lets Devin draw GPT model usage from any personal ChatGPT plan (Go, Plus or Pro), widening the Sep 29 'Sign in with ChatGPT' launch that covered Plus and Pro. Once linked under Settings > Connections and the 'Use your ChatGPT plan' switch is on, the OpenAI portion of Devin Cloud, Desktop and CLI sessions bills to the ChatGPT subscription (counting toward its Codex/ChatGPT Work usage) and never against Devin quota; other models in the mix still use Devin quota. Cloud modes lean more on OpenAI models when a plan is linked. One toggle covers all three surfaces, Teams members link individually, fast/priority variants are excluded, and sessions fall back to Devin quota if the ChatGPT allowance runs out or the grant expires. Available on self-serve Devin Pro, Max and Teams only. Cognition cites Artificial Analysis Coding Agent Index v1.5 (vendor-reported): Fusion with GPT-6 Astra lead plus SWE-2 sidekick scores 58.9 at $4.54 per task versus Codex Astra (max) at 61.6 and $7.47, about 39% cheaper.

Action HubReady to adopt this in production?

Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.