GPT-5.5
OpenAI · released Apr 24, 2026
- Type
- Closed
- Input
- $5.00
- Output
- $30.00
- Cached
- $0.50
List price · per 1M tokens · OpenAI at 1.1M context · machine-readable source ↗
Our take
Written Sep 30, 2026GPT-5.5 is a hosted model with no download, so using it means choosing a host. Its measured results are strongest on data analysis and set-piece code, and weakest when an agent has to finish a multi-step task on its own.
Reach for it on table and event-ordering work, where it places 2nd of 58 on LiveBench Data Analysis as of 25 Jun 2026, and on set-piece code generation, where it places 8th of 58 on LiveBench Coding as of 25 Jun 2026. Long documents need not be split up first, and it accepts text, images and files. Skip it if you need fixes landed in an existing codebase without review, or if you need an agent to finish a multi-step task on its own.
The case for it
- 2nd of 58 on LiveBench Data Analysis as of 25 Jun 2026, a board of table and event-ordering tasks, so the result covers structured-data work rather than open-ended reasoning.
- 8th of 58 on LiveBench Coding as of 25 Jun 2026, which covers code generation and completion rather than fixing issues in an existing project.
- 2nd of 55 on Arena Agent · Tool use as of 25 Sep 2026, a score for calling the right tool and not inventing one, so it reflects tool selection rather than whether the task was finished.
- 5th of 42 on SWE-bench Verified as of 19 Feb 2026, a result for the model inside that harness, measuring the share of real GitHub issues resolved end-to-end.
The case against it
- 27th of 58 on LiveBench Agentic Coding as of 25 Jun 2026, a board run inside an agent harness, so it trails its own 8th of 58 on LiveBench Coding.
- 35th of 55 on Arena Agent · Task outcome as of 25 Sep 2026, a score for finishing the task the session set out to do, against its 2nd of 55 on Arena Agent · Tool use.
- We list no download for it, so using it means choosing a host, and the cheapest listed rate sits well under what the other hosts charge.
How good is it?
A general-purpose text model for everyday questions, drafting and coding, and for calling tools to carry out requests.
- getting answers to everyday questionsArena Text (overall) · 20th of 168
- drafts, rewrites and editingArena Creative Writing · 20th of 168
- writing and completing codeArena Coding · 32nd of 168
- calling tools to carry out requestsArena Agent · Tool use · 2nd of 55
EverydayGeneral questions and everyday reasoning
Arena Text (overall)20th of 168 · 1477
Also on this board: 1481 (Sep 25, 2026). Read the pair, not the higher one.
CodingWriting and fixing code on its own
Arena Coding32nd of 168 · 1513
Also on this board: 1519 (Sep 25, 2026). Read the pair, not the higher one.
AgenticPlanning, calling tools, staying on task
Arena Agent20th of 55 · 0.018
Also on this board: 0.056 (Sep 5, 2026), 0.044 (Sep 25, 2026). Read the pair, not the higher one.
WritingDrafting and rewriting prose
Arena Creative Writing20th of 168 · 1456
Also on this board: 1452 (Sep 25, 2026). Read the pair, not the higher one.
Placings on Arena's agent boards, from live sessions people ran themselves. A model can lead on one of these and sit mid-field on the others.
Boards this model appears on that none of the ratings above are built on.
Every published score for this model21 scoresEvery figure we hold, from 21 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 between 59 min and 1 hour ago — each listing carries its own date.
OpenAI, direct
The lab is the cheapest at this context. The strip above and this offer are the same one, compared at 1.1M of context. One cheaper row below is outside that comparison: a non-standard pricing tier.
- per 1M tokens
- $5.00 in / $30.00 out
- Context served
- 1.1M
- Throughput
- ~39 tok/s
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenAIflex tierThrough OpenRouter | $2.50 / $15.00checked 1 hour ago | 1.1M128K max reply | 99 tok/s | No | Yesunknown period | Unknown |
| OpenRouterOpenRouter's own listing | $5.00 / $30.00checked 1 hour ago | 1.1M | not measured | Unknown | Unknown | Unknown |
| OpenAIDirect | $5.00 / $30.00checked 59 min ago | 1.1M128K max reply | 39 tok/s | No | Yesunknown period | Unknown |
| Microsoft Azure AIThrough OpenRouter | $5.00 / $30.00checked 1 hour ago | 1.1M128K max reply | 63 tok/s | No | No | Confirmed |
| Amazon Bedrockus-east-1Through OpenRouter | $5.50 / $33.00checked 1 hour ago | 1.1M128K max reply | 92 tok/s | No | No | Unknown |
| Microsoft Azure AIeuThrough OpenRouter | $5.50 / $33.00checked 1 hour ago | 1.1M128K max reply | 30 tok/s | No | No | Confirmed |
| Microsoft Azure AIusThrough OpenRouter | $5.50 / $33.00checked 1 hour ago | 1.1M128K max reply | 34 tok/s | No | No | Confirmed |
| OpenAIfast tierThrough OpenRouter | $12.50 / $75.00checked 1 hour ago | 1.1M128K max reply | 80 tok/s | No | Yesunknown period | Unknown |
OpenRouter
Microsoft Azure AI
Amazon BedrockAcross the 8 listings we hold: 7 say they do not train on prompts, 0 say they do and 1 does not say. 3 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 | ✓ | ✓ | ✓ |
| OpenAIDirect | ✓ | ✓ | ✓ |
| Microsoft Azure AIThrough OpenRouter | ✓ | ✓ | ✓ |
| Amazon Bedrockus-east-1Through OpenRouter | ✓ | ✓ | ✓ |
| Microsoft Azure AIeuThrough OpenRouter | ✓ | ✓ | ✓ |
| Microsoft Azure AIusThrough OpenRouter | ✓ | ✓ | ✓ |
| OpenAIfastThrough OpenRouter | ✓ | ✓ | ✓ |
Tool calling: 8 of 8 listings say yes. JSON output: 8 of 8 listings say yes. Strict schema: 8 of 8 listings say yes.
Models people weigh against GPT-5.5
When we formed this view
Recent changes
Each 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 8 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-5-5