GPT-6 Astra in OpenAI Codex
OpenAI - Powerful
In OpenAI Codex, GPT-6 Astra meters at 250 credits per 1M input tokens and 1250 per 1M output tokens. At $0.04 per credit that is exactly $10.00 and $50.00 per 1M tokens, the same as calling the OpenAI API directly. A typical daily-coding request works out to about 3.5 credits (about $0.14). Verified against OpenAI's Codex rate card on 2026-08-22.
Credit rate card
- Credits / 1M in
- 250
- Credits / 1M cached
- 25
- Credits / 1M out
- 1,250
- Cost per credit
- $0.04
Credits per request
Defined workload baskets, not measured tasks.
| Workload | Credits | Cost | Direct API cost |
|---|---|---|---|
| Short chat turn | 1.8 | $0.07 | $0.07 |
| Daily coding request | 3.5 | $0.14 | $0.14 |
| Agent session step | 11.7 | $0.47 | $0.47 |
At about 3.5 credits per daily-coding request, GPT-6 Astra ranks #5 of 5 current-generation models by cost, about 45.2x the cheapest (GPT-5.6 Luna).
Credit-to-API parity
The implied dollar value of a credit spent on GPT-6 Astra, backed out from the direct API rate card divided by the credit rate card. It should land on $0.04 for both input and output.
- Implied $ / credit, input
- $0.04
- Implied $ / credit, output
- $0.04
- Markup at the daily workload
- 0.00%
- Ratio match
- Exact
Input : cached : output ratio in credits is 10 : 1 : 50, versus 10 : 1 : 50 in direct API dollars.