GPT-5.4 Nano
gpt-5.4-nano
Multimodal
6 gateways
1.62x price spread
Where you can run it
| Gateway | Its id | Context | Max output | Input / Mtok | Output / Mtok |
|---|---|---|---|---|---|
| Cheaper Inference | gpt-5.4-nano |
400K | 128K | $0.132 | $0.827 |
| Concentrate | gpt-5.4-nano |
400K | 128K | — | — |
| Experiential Labs | gpt-5.4-nano |
400K | 128K | $0.200 | $1.25 |
| LLM Gateway | gpt-5.4-nano |
400K | — | $0.200 | $1.25 |
| Ramp Router | gpt-5.4-nano |
400K | 128K | $0.200 | $1.25 |
| SayGM Cheapest input | gpt-5.4-nano |
— | — | $0.124 | $0.772 |
Prices are US dollars per million tokens, as published in each gateway's own catalog at the last crawl. A dash means that gateway publishes no value — not that it is free. Rows differ because gateways set their own markup, and some let you bring your own provider key.
- Canonical id
gpt-5.4-nano- Vendor
- OpenAI
- Type
- Multimodal
- Gateways serving it
- Cheaper InferenceConcentrateExperiential LabsLLM GatewayRampSayGM
- Largest context window
- 400K tokens
- Input price range
- $0.124 – $0.200 / Mtok
- Price spread
- 1.62x between cheapest and dearest gateway
- Accepts
- Text, Image
- Returns
- Text
- Cataloged since
History
What changed for GPT-5.4 Nano
Recorded by the daily catalog crawl — when gateways picked this model up, dropped it, or repriced it.