GPT-5.4
OpenAI's latest frontier model, unifying the Codex and GPT lines. Features 1M+ token context window with strong coding, document understanding, and tool use.
- Input price
- $2.25–$4.50list $2.50-10%
- Output price
- $13.50–$20.25list $15.00-10%
- Context
- 1.1M
- Max output
- 128K
- Modalities
- InputTextImageOutputText
- Capability
- 157 (#22 of 89)
- Human preference
- 1452
- 30-day uptime
- —
- Released
- 2026-03-05
- Knowledge cutoff
- 2025-08
Providers
This model is served by the dedicated providers below. We pick the provider for each request, so provider preferences don’t apply, including a provider named in your request. Prices in USD per 1M tokens. Need a custom setup? Contact [email protected].
- SuccessTTFT
- Azure AI FoundryGlobal——
- OpenAIUnited States——
Pricing
Prices for this model. Each request is billed at these prices, whichever provider serves it.
| Item | Standard | Flex | Batch | Priority | Price |
|---|---|---|---|---|---|
| Prompt ≤ 272K tokens | |||||
| Inputper 1M tokens | $2.25list $2.50-10% | $1.125list $1.25-10% | $1.125list $1.25-10% | $4.50list $5.00-10% | — |
| Outputper 1M tokens | $13.50list $15.00-10% | $6.75list $7.50-10% | $6.75list $7.50-10% | $27.00list $30.00-10% | — |
| Cache readper 1M tokens | $0.225list $0.25-10% | $0.117list $0.13-10% | $0.117list $0.13-10% | $0.45list $0.50-10% | — |
| Prompt > 272K tokens | |||||
| Inputper 1M tokens | $4.50list $5.00-10% | $2.25list $2.50-10% | $2.25list $2.50-10% | — | — |
| Outputper 1M tokens | $20.25list $22.50-10% | $10.125list $11.25-10% | $10.125list $11.25-10% | — | — |
| Cache readper 1M tokens | $0.45list $0.50-10% | $0.225list $0.25-10% | $0.225list $0.25-10% | — | — |
| Web searchper request | — | — | — | — | $0.009list $0.01-10% |
Make a request
Set your provider preferences, then copy the request. It reads your API key from OHMYGPT_API_KEY.
Similar models
Other models with the same output types at a similar price.
Human preference ratings are from the LMArena Text Arena (CC BY 4.0, as of 2026-10-28): Elo ratings from blind votes, on a different scale from capability scores. lmarena.ai/leaderboard/text