GPT-5.5
OpenAI · released Apr 24, 2026
- Type
- Proprietary
- Input
- $5.00
- Output
- $30.00
- Cached
- $0.50
List price · per 1M tokens · OpenAI at 1.1M context · source ↗
Our take
Written Aug 3, 2026GPT-5.5 is OpenAI's flagship multimodal model that can handle up to 1.05 million tokens in a single request, making it one of the longest-context frontier offerings available. It excels at coding tasks but scores lowest on creative writing among its measured capabilities, with tiered pricing that spans a wide range.
Choose this for analysing full book-length documents or large codebases in one pass, or for premium coding assistance where it scores highest in its benchmark suite. Use it for enterprise deployments through Azure or Amazon Bedrock when compliance requirements matter. Skip it if you need strong creative writing output, want to self-host or fine-tune, or are looking for a budget option under a dollar per million tokens.
The case for it
- One of the longest context windows among frontier-class models we track, at 1.05 million tokens.
- Top measured coding performance in its benchmark suite, with an Arena Coding Elo of 1520.4661.
- Wide provider choice with a tenfold price spread between base and premium tiers for budget flexibility.
- Hard-prompt handling is notably stronger than its creative writing, with a 44.9314-point Elo gap between the two.
The case against it
- Creative writing is its weakest measured capability, scoring lowest in its entire benchmark set.
- Expensive at every tier, with no sub-dollar offering available.
- Throughput is inconsistent and sometimes slow for the price, with the cheaper Azure tier delivering 1.84 times the speed of a more expensive OpenAI tier.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)11th of 143 · 1476.3
Also on this board: 1482.4 via High (Aug 2, 2026). Read the pair, not the higher one.
CodingWriting and fixing code on its own
Arena Coding20th of 143 · 1514.3
Also on this board: 1520.5 via High (Aug 2, 2026). Read the pair, not the higher one.
AgenticPlanning, calling tools, staying on task
Arena Agent (IPS)10th of 36 · 0.055
Also on this board: 0.079 via High (Jul 28, 2026), 0.085 via xHigh (Jul 28, 2026). Read the pair, not the higher one.
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-5.5 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 model17 scoresEvery figure we hold, from 17 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
The lab is also the cheapest we hold. The strip above and this offer are the same one, so nothing on this page undercuts OpenAI on 1.1M of context.
- per 1M tokens
- $5.00 in / $30.00 out
- Context served
- 1.1M
- Throughput
- ~43 tok/s
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenAIdefault tier | $1.25 / $10.00 | 400K | 40 tok/s | No | Yesunknown period | Unknown |
| Microsoft Azure AI | $1.25 / $10.00 | 400K | 68 tok/s | No | No | Confirmed |
| Microsoft Azure AIswedencentral | $1.38 / $11.00 | 400K | 83 tok/s | No | No | Confirmed |
| OpenAIflex tier | $2.50 / $15.00 | 1.1M | 82 tok/s | No | Yesunknown period | Unknown |
| OpenAI | $5.00 / $30.00 | 1.1M128K out | 43 tok/s | No | Yesunknown period | Unknown |
| OpenRouter | $5.00 / $30.00 | 1.1M | not measured | Unknown | Unknown | Unknown |
| Amazon Bedrockus-east-1 | $5.50 / $33.00 | 1.1M | 52 tok/s | No | No | Unknown |
| Microsoft Azure AIeu | $5.50 / $33.00 | 1.1M | 75 tok/s | No | No | Confirmed |
| OpenAIpriority tier | $12.50 / $75.00 | 1.1M | 50 tok/s | No | Yesunknown period | Unknown |
Microsoft Azure AI
OpenRouter
Amazon BedrockAcross the 9 listings we hold: 8 say they do not train on prompts, 0 say they do and 1 do not say. 3 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 |
|---|---|---|---|
| OpenAIdefault | ✓ | ✓ | ✓ |
| Microsoft Azure AI | ✓ | ✓ | ✓ |
| Microsoft Azure AIswedencentral | ✓ | ✓ | ✓ |
| OpenAIflex | ✓ | ✓ | ✓ |
| OpenAI | ✓ | ✓ | ✓ |
| OpenRouter | ✓ | ✓ | ✓ |
| Amazon Bedrockus-east-1 | ✓ | ✗ | ✗ |
| Microsoft Azure AIeu | ✓ | ✓ | ✓ |
| OpenAIpriority | ✓ | ✓ | ✓ |
Tool calling: 9 of 9 listings say yes. JSON output: 8 of 9 listings say yes, 1 says no. Strict schema: 8 of 9 listings say yes, 1 says no.
Models people weigh against GPT-5.5
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 9 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-5-5