GPT-6.2 Astra & Sol vs MAI-Code-2 in Copilot and Codex
Both OpenAI's GPT-6.2 Astra / Sol and Microsoft's MAI-Code-2 / Flash rolled out to GitHub Copilot on November 17, 2026 with 1.05M token context windows—but their plan gates differ sharply. In GitHub Copilot, GPT-6.2 Astra and Sol require Pro+, Max, Business, or Enterprise (usage-based billing), whereas MAI Code 2 Flash unlocks on Copilot Free and Student and MAI Code 2 starts at Copilot Pro. Inside OpenAI Codex (app & CLI), GPT-6.2 Astra and Sol run on ChatGPT Plus through Enterprise/Edu, while MAI-Code-2 is Microsoft-hosted only.
Compare Copilot & Codex availability ↓1. Copilot, Codex & API pricing comparison
| Dimension | OpenAI GPT-6.2 Astra & Sol | Microsoft MAI-Code-2 & Flash |
|---|---|---|
| GitHub Copilot plan tiers | Copilot Pro+, Max, Business, and Enterprise only (usage-based billing at provider list pricing; excludes Free, Student, and base Pro). | MAI Code 2: Copilot Pro, Pro+, Max, Business, and Enterprise. MAI Code 2 Flash: Copilot Free, Student, Pro, Pro+, Max, Business, and Enterprise. |
| GitHub Copilot surfaces | VS Code, Visual Studio, Copilot CLI, GitHub Copilot coding agent, Copilot app, github.com, GitHub Mobile (iOS/Android), JetBrains IDEs, Xcode, Eclipse. | VS Code, Visual Studio, Copilot CLI, GitHub Copilot cloud agent, Copilot app, github.com, GitHub Mobile (iOS/Android), JetBrains, Xcode, Eclipse. |
| OpenAI Codex (App & CLI) | Available natively in Codex desktop app and CLI across ChatGPT Plus, Pro, Business, Enterprise, and Edu plans (Ultrafast on Pro $200 for Sol; Pro $500 & Enterprise/Edu for both). | Not available in OpenAI Codex. MAI Code 2 and MAI Code 2 Flash are Microsoft AI models delivered via GitHub Copilot and Microsoft Foundry. |
| API token pricing (per 1M) | Astra: $8.00 in / $0.40 cached / $40.00 out. Sol: $2.00 in / $0.10 cached / $10.00 out. (>272K input doubles input/cache & 1.5x output.) | MAI Code 2 promo (until May 17, 2027): $0.50 in / $2.50 out (list $1.00 / $5.00). MAI Code 2 Flash targets routine edits at lower cost. |
| Context & output limits | 1,050,000-token context window / 128,000 max output tokens (with 272K long-context billing step). | 1,050,000-token context window (922,000 input / 128,000 output). |
| Primary engineering fit | Astra: multi-service bug tracing, terminal research, hard reasoning. Sol: in-depth PR review (~2x more real issues caught vs GPT-6.1 Sol) & daily automation. | MAI Code 2: multi-file implementation, deep debugging, repo-wide reasoning. MAI Code 2 Flash: targeted edits, explanations & routine CI fixes. |
2. Admin policies, IDEs & cost traps to watch
Enterprise & Business enablement: On GitHub Copilot Business and Enterprise, both GPT-6.2 and MAI-Code-2 models default to enabled unless an administrator has disabled the global default policy or explicitly restricted the model in Copilot settings. In ChatGPT Work / Codex Enterprise and Edu workspaces, admins must explicitly enable GPT-6.2 Astra and Sol before team members can select them.
Long-context pricing vs. flat 1.05M window: While both families expose ~1.05M context windows, OpenAI doubles input/cache rates and applies a 1.5x output multiplier whenever a GPT-6.2 request exceeds 272,000 input tokens. Keep Codex or CI diff prompts scoped cleanly, and track subscription quotas with Claude Code vs Codex usage trackers.
Frequently asked questions
Can Copilot Free, Student, or base Pro users select GPT-6.2 Astra or Sol?
No. According to the November 17, 2026 GitHub Changelog, OpenAI GPT-6.2 Astra and GPT-6.2 Sol in GitHub Copilot are restricted to Copilot Pro+, Max, Business, and Enterprise plans under usage-based billing.
Which coding model is available on GitHub Copilot Free and Student plans?
Microsoft MAI Code 2 Flash is available across Copilot Free, Student, Pro, Pro+, Max, Business, and Enterprise. Standard MAI Code 2 starts at Copilot Pro and above.
Can I run Microsoft MAI-Code-2 inside OpenAI Codex CLI or desktop app?
No. MAI-Code-2 and MAI-Code-2 Flash are Microsoft AI models hosted in GitHub Copilot and Microsoft Foundry; OpenAI Codex app and CLI run OpenAI models such as GPT-6.2 Astra and GPT-6.2 Sol.
How do MAI-Code-2 API rates compare to GPT-6.2 Sol and Astra?
During its promotional period through May 17, 2027, MAI Code 2 is $0.50 input / $2.50 output per 1M tokens (moving to $1.00 / $5.00 list), compared to $2.00 / $10.00 for GPT-6.2 Sol and $8.00 / $40.00 for GPT-6.2 Astra.