GPT-6.2 Astra in GitHub Copilot
OpenAI - Powerful
In GitHub Copilot, GPT-6.2 Astra costs $8 per 1M input tokens and $40 per 1M output tokens, billed as AI credits at $0.01 each. A typical daily-coding request works out to about 9.9 credits (about $0.10) with auto model selection on. Rates verified against GitHub's rate sheet on 2026-11-20.
Token rates
- Input / 1M
- $8.00
- Cached input / 1M
- $0.40
- Output / 1M
- $40.00
- Cost per credit
- $0.01
Cache writes bill at $10.00/1M tokens (Anthropic models).
Requests above 272,000 input tokens bill at the long-context tier ($16.00 in / $60.00 out per 1M).
Credits per request
Defined workload baskets, not measured tasks, with auto model selection on.
| Workload | Credits (auto on) | Credits (auto off) | Cost |
|---|---|---|---|
| Short chat turn | 5.1 | 5.7 | $0.05 |
| Daily coding request | 9.9 | 11.0 | $0.10 |
| Agent session step | 32.6 | 36.2 | $0.33 |
At about 9.9 credits per daily-coding request, GPT-6.2 Astra ranks #30 of 34 current models by cost, about 156.6x the cheapest (MAI Code 2 Flash).
Requests per plan allowance
Daily-coding requests the included monthly credits buy, before any overage.
| Plan | Credits / month | Requests |
|---|---|---|
| Copilot Pro | 1,500 | 150 |
| Copilot Pro+ | 7,000 | 704 |
| Copilot Max | 20,000 | 2,012 |
| Copilot Business | 1,900 | 191 |
| Copilot Enterprise | 3,900 | 392 |