Codex pricing and limits
Codex comes with every ChatGPT plan, and its usage is shared with ChatGPT and limited per five hours on Plus. The other way to pay is an API key. Here is what each costs, and how to run Codex CLI on any model with one key.
Competitor details were checked against their official pages on November 4, 2026. Prices and limits change; follow the links to confirm.
Codex on a ChatGPT plan
As of November 4, 2026, OpenAI's Codex pricing page lists:
| Plan | Price | Codex on it |
|---|---|---|
| Free | $0 a month | Quick coding tasks; GPT-6 Luna at standard speed in the desktop app, subject to rollout |
| Go | $8 a month | Lightweight coding tasks |
| Plus | $20 a month | Codex on the web, in the CLI, the IDE extension and iOS; GPT-6.1 Sol and GPT-6 Luna |
| Pro | From $100 a month ($100, $200 or $500) | Everything in Plus; currently no five-hour limit |
| Business | $20 per user a month, billed annually ($25 monthly) | Codex across the team's workspace |
Two things matter for heavy use. ChatGPT and Codex share one allowance, so chat usage counts against coding. And on Plus, usage is estimated "per five-hour period", with the estimate depending on the model, the task, and how much context it holds.
Codex on an API key
OpenAI's page lists a second option: sign Codex in with an API key and pay for usage at API pricing. You get Codex in the CLI, SDK and IDE extension (not the cloud features), and no plan windows — you pay for the tokens you use.
An Inferray key works there too, and opens the rest of the catalog to Codex:
model = "gpt-6.1-sol"
model_provider = "inferray"
[model_providers.inferray]
name = "Inferray"
base_url = "https://api.inferray.com/v1"
env_key = "INFERRAY_API_KEY"
wire_api = "responses"
The Codex setup guide has the details. Switch models per run with codex --model claude-opus-5-5.
| Model | Input / 1M | Cached / 1M | Output / 1M | Context | Token Plans |
|---|---|---|---|---|---|
| GPT-6.1 Sol | $2 | $0.1 | $10 | 1.1M | Included |
| GPT-6 Luna | $0.1 | $0.01 | $0.5 | 1.1M | Included |
| Claude Opus 5.5 | $4 | $0.2 | $20 | 1M | Included |
| Claude Sonnet 5.5 | $2 | $0.1 | $10 | 1M | Included |
| Gemini 4 Argon | $4 | $0.2 | $20 | 1.0M | Included |
| GLM-5.3 | $0.77 | $0.143 | $2.42 | 262K | Included |
Or a Token Plan
| Plan | You pay / month | Usage at list price | ≈ tokens on GPT-6.1 Sol | ≈ tokens on Claude Sonnet 5.5 | ≈ tokens on GLM-5.3 |
|---|---|---|---|---|---|
| Solo | $20 | $60 | 96M | 96M | 210M |
| Pro | $100 | $300 | 480M | 480M | 1B |
| Max | $200 | $600 | 960M | 960M | 2.1B |
Token estimates assume coding-agent traffic: 85% cached input, 12% fresh input, 3% output.
A plan key is an ordinary API key, so Codex takes it the same way. No five-hour windows, no weekly cap, and the credit isn't shared with a chat app.
Prices on this page come from the model catalog and were current on November 12, 2026.