GPT-5.4 Mini
OpenAI · released Mar 17, 2026
- Type
- Proprietary
- Input
- $0.75
- Output
- $4.50
- Cached
- $0.075
List price · per 1M tokens · OpenAI at 272K context · source ↗
Our take
Written Aug 3, 2026GPT-5.4 Mini is a compact hosted model from OpenAI that handles text, images and files with a 400,000-token working memory. It offers tiered throughput pricing and has been measured across more Arena categories than any other compact model in the OpenAI line, though its creative writing and web-development scores lag its own coding performance.
Pick this for high-throughput batch processing where speed matters more than peak quality, or for long-context document and image analysis that needs 400,000 tokens of working memory. Use it for coding assistance where its Arena Coding score is the relevant signal. Skip it if you need strong creative writing, web-development coding, or if you want to run the model on your own hardware rather than through an API.
The case for it
- Widest benchmark coverage in the compact OpenAI line, with seven distinct Arena categories measured.
- Exceptional context length for its price tier: 400,000 tokens.
- Throughput can be traded for cost on the same provider, with a fivefold price spread between the cheapest and mid tiers.
The case against it
- Web development coding lags its other coding skills by nearly a hundred Elo points.
- Creative writing is its weakest measured category, below even its own web-development scores.
- No parameter transparency: total and active parameters are both undisclosed, so the 400,000-token context cannot be assessed for efficiency.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)37th of 143 · 1448.6via High
CodingWriting and fixing code on its own
Arena Coding40th of 143 · 1497via High
AgenticPlanning, calling tools, staying on task
GPT-5.4 Mini is not on Arena Agent (IPS), which is where the rating would come from, so there is no rating here. It is on LiveBench Agentic Coding, in 29th of 35 with 41.7.
WritingWe do not rate this
Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage. So we show where GPT-5.4 Mini placed and give it no mark out of five.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.
Every published score for this model15 scoresEvery figure we hold, from 15 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Or rent it from someone else
Why this differs from the header. The strip above quotes OpenAI's own list price. This is the cheapest live offer at the widest standard context we hold, whoever is serving it — a reseller undercutting a lab is ordinary commerce, not an error.
- per 1M tokens
- $0.75 in / $4.50 out
- Context served
- 400K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenAIflex tier | $0.38 / $2.25 | 400K | 221 tok/s | No | Yesunknown period | Unknown |
| OpenRouter | $0.75 / $4.50 | 400K | not measured | Unknown | Unknown | Unknown |
| Microsoft Azure AI | $0.75 / $4.50 | 400K | 63 tok/s | No | No | Confirmed |
| OpenAI | $0.75 / $4.50 | 272K128K out | 41 tok/s | No | Yesunknown period | Unknown |
| OpenAIpriority tier | $1.50 / $9.00 | 400K | 74 tok/s | No | Yesunknown period | Unknown |
Microsoft Azure AIAcross the 5 listings we hold: 4 say they do not train on prompts, 0 say they do and 1 do not say. 1 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenAIflex | ✓ | ✓ | ✓ |
| OpenRouter | ✓ | ✓ | ✓ |
| Microsoft Azure AI | ✓ | ✓ | ✓ |
| OpenAI | ✓ | ✓ | ✓ |
| OpenAIpriority | ✓ | ✓ | ✓ |
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 Mini
When we formed this view
Dates behind this page
Prices last checked 9d ago
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 do not say whether they train on prompts.
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
Commercial API terms. We hold no licence record for this model, so there is nothing to summarise here.
Identifiers
- Modality record
- text+image+file->text
- Catalogue slug
- openai-gpt-5-4-mini