GPT-5.4 Nano
OpenAI · released Mar 17, 2026
- Type
- Closed
- Input
- $0.20
- Output
- $1.25
- Cached
- $0.020
List price · per 1M tokens · OpenAI at 272K context · machine-readable source ↗
Our take
Written Sep 30, 2026GPT-5.4 Nano is a hosted-only model with a low entry price and support for images and files alongside the question. Its measured results are middling to poor, so it suits high-volume work where the bill matters more than peak quality.
Reach for it on high-volume, cost-sensitive work where a mid-table model is good enough, or when your prompts carry images or files alongside the question. It is cheap to reach through a host and takes long documents without splitting them up first. Skip it if you need strong language comprehension or coding, or if you meant to run the model on your own machine.
The case for it
- The cheapest listed offer is well under the next tier up, so the entry price is low for a model that also takes images and files.
- Long documents fit beside the question without being broken apart first, though reliable recall across all of it is unverified in our data.
- Competition-style maths is its best measured category, at 90.98% on LiveBench Mathematics, 26th of 58 on LiveBench Mathematics via xHigh as of 25 Jun 2026.
The case against it
- Language comprehension is its weakest measured area: 62.51% on LiveBench Language, 58th of 58 on LiveBench Language via xHigh as of 25 Jun 2026, last in the field.
- Coding sits below the middle of the field, at 50th of 58 on LiveBench Coding via xHigh as of 25 Jun 2026, with agentic coding at 43rd of 58 on the same board and date.
- We list no download for it, so using it means choosing a host; there is no way to run it yourself.
How good is it?
EverydayGeneral questions and everyday reasoning
Arena Text (overall)95th of 168 · 1401
CodingWriting and fixing code on its own
Arena Coding84th of 168 · 1460
AgenticPlanning, calling tools, staying on task
Not yet scored on Arena Agent. It is on LiveBench Agentic Coding, in 43rd of 58 with 46.77.
WritingDrafting and rewriting prose
Arena Creative Writing110th of 168 · 1335
Boards this model appears on that none of the ratings above are built on.
Every published score for this model14 scoresEvery figure we hold, from 14 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Where to rent it
Prices checked 4 hours ago — each listing carries its own date.
Microsoft Azure AI, through OpenRouter
Why this differs from the header. The strip above quotes OpenAI's own list price; this is the cheapest live offer, whoever is serving it — a reseller undercutting a lab is ordinary commerce, not an error.
- per 1M tokens
- $0.20 in / $1.25 out
- Context served
- 400K
- Throughput
- ~43 tok/s
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenAIflex tierThrough OpenRouter | $0.10 / $0.63checked 4 hours ago | 400K128K max reply | 134 tok/s | No | Yesunknown period | Unknown |
| OpenRouterOpenRouter's own listing | $0.20 / $1.25checked 4 hours ago | 400K | not measured | Unknown | Unknown | Unknown |
| Microsoft Azure AIThrough OpenRouter | $0.20 / $1.25checked 4 hours ago | 400K128K max reply | 43 tok/s | No | No | Confirmed |
| OpenAIDirect | $0.20 / $1.25checked 4 hours ago | 272K128K max reply | 74 tok/s | No | Yesunknown period | Unknown |
| Microsoft Azure AIusThrough OpenRouter | $0.22 / $1.38checked 4 hours ago | 400K128K max reply | 11 tok/s | No | No | Confirmed |
OpenRouterAcross the 5 listings we hold: 4 say they do not train on prompts, 0 say they do and 1 does not say. 2 appear in the zero-retention registry we check; the rest are unknown to us.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenAIflexThrough OpenRouter | ✓ | ✓ | ✓ |
| OpenRouterOpenRouter's own listing | ✓ | ✓ | ✓ |
| Microsoft Azure AIThrough OpenRouter | ✓ | ✓ | ✓ |
| OpenAIDirect | ✓ | ✓ | ✓ |
| Microsoft Azure AIusThrough OpenRouter | ✓ | ✓ | ✓ |
Tool calling: 5 of 5 listings say yes. JSON output: 5 of 5 listings say yes. Strict schema: 5 of 5 listings say yes.
Models people weigh against GPT-5.4 Nano
When we formed this view
Recent changes
What moved
first indexed by our pipelineEach date is the day we first saw the change, or the day the maker announced it.
What we do not know about this model yet
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 5 listings does not say whether it trains on prompts.
- We hold no batch or off-peak rate for any of its listings.
Licence and identifiers
What the licence allowsWe hold no licence record for this model. Inside are the identifiers you need to pull it — its Hugging Face repo where we have one, our slug and a machine-readable card.
Licence
We hold no licence record for this model, and no record of published weights either — so we can neither summarise its terms nor point you at the weights.
Identifiers
- Takes in, gives back
- Text, images and documents in, text out
- Catalogue slug
- openai-gpt-5-4-nano