OpenAI model product
GPT-5.5
5 is an OpenAI large language model for agentic coding, computer use, knowledge work, and early scientific research.
Updated Sep 8, 2026. Default version: GPT-5.5
LLMBoard Score81.7GPT-5.5
Coverage40%26 benchmark families
Context window1.1MTokens
Official input price$5OpenAI API
GPT-5.5 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
GPT-5.5 Benchmark Results
Benchmark scores for GPT-5.5.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkBixBench | Score80.50% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkGDPval-MM | Score84.90% | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 2.0 | Score82.70% | Rank01 | Participants51 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkARC-AGI | Score95.00% | Rank02 | Participants9 | Percentile87.50% | EvidenceC | Evaluated |
| BenchmarkARC-AGI v2 | Score85.00% | Rank02 | Participants18 | Percentile94.12% | EvidenceC | Evaluated |
| BenchmarkGeneBench | Score25.00% | Rank02 | Participants2 | Percentile0.00% | EvidenceC | Evaluated |
| BenchmarkLiveBench | Score80.71% | Rank02 | Participants38 | Percentile97.30% | EvidenceB | Evaluated |
| BenchmarkMMMU-Pro | Score83.20% | Rank02 | Participants69 | Percentile98.53% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Bash Recovery Steps | Score13.25% | Rank02 | Participants49 | Percentile97.92% | EvidenceA | Evaluated |
| BenchmarkLM Arena Search Factuality | Score1,245.34 rating | Rank02 | Participants28 | Percentile96.30% | EvidenceA | Evaluated |
| BenchmarkGraphwalks parents >128k | Score58.50% | Rank03 | Participants7 | Percentile66.67% | EvidenceC | Evaluated |
| BenchmarkLM Arena Search | Score1,242.05 rating | Rank03 | Participants28 | Percentile92.59% | EvidenceA | Evaluated |
| BenchmarkLM Arena Search Style Control | Score1,223.60 rating | Rank03 | Participants28 | Percentile92.59% | EvidenceA | Evaluated |
| BenchmarkFinance Agent | Score60.00% | Rank04 | Participants8 | Percentile57.14% | EvidenceC | Evaluated |
| BenchmarkFinance Agent v2 | Score51.76% | Rank05 | Participants26 | Percentile84.00% | EvidenceB | Evaluated |
| BenchmarkLM Arena Document | Score1,484.70 rating | Rank05 | Participants33 | Percentile87.50% | EvidenceA | Evaluated |
| BenchmarkCyberGym | Score81.80% | Rank06 | Participants15 | Percentile64.29% | EvidenceC | Evaluated |
| BenchmarkTau2 Telecom | Score98.00% | Rank06 | Participants36 | Percentile85.71% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Steerability | Score8.54% | Rank06 | Participants49 | Percentile89.58% | EvidenceA | Evaluated |
| BenchmarkFrontierCode 1.1 | Score43.00% | Rank07 | Participants17 | Percentile62.50% | EvidenceB | Evaluated |
| BenchmarkFrontierMath | Score35.40% | Rank07 | Participants17 | Percentile62.50% | EvidenceC | Evaluated |
| BenchmarkFrontierSWE | Score73.00% | Rank07 | Participants16 | Percentile60.00% | EvidenceB | Evaluated |
| BenchmarkGPQA | Score93.60% | Rank07 | Participants247 | Percentile97.56% | EvidenceC | Evaluated |
| BenchmarkGraphwalks BFS >128k | Score45.40% | Rank07 | Participants11 | Percentile40.00% | EvidenceC | Evaluated |
| BenchmarkLegal Agent Benchmark | Score2.10% | Rank07 | Participants13 | Percentile50.00% | EvidenceB | Evaluated |
| BenchmarkMRCR v2 (8-needle) | Score74.00% | Rank07 | Participants24 | Percentile73.91% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Tool Hallucination | Score0.78% | Rank07 | Participants49 | Percentile87.50% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Factuality | Score1,485.74 rating | Rank07 | Participants121 | Percentile95.00% | EvidenceA | Evaluated |
| BenchmarkOfficeQA Pro | Score54.10% | Rank08 | Participants9 | Percentile12.50% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Leaderboard | Score7.21% | Rank08 | Participants49 | Percentile85.42% | EvidenceA | Evaluated |
GPT-5.5 Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext factuality | Categoryjapanese | Rank01 | Rating / score1,504.93 | Votes510 | ObservationsN/A | Result date |
| Arenaagent bash recovery steps | Categoryoverall | Rank02 | Rating / score0.13 | VotesN/A | Observations43.3K | Result date |
| Arenasearch factuality | Categoryoverall | Rank02 | Rating / score1,245.34 | Votes89,833 | ObservationsN/A | Result date |
| Arenatext factuality | Categorygerman | Rank02 | Rating / score1,474.75 | Votes542 | ObservationsN/A | Result date |
| Arenavision style control | Categoryhomework | Rank02 | Rating / score1,338.55 | Votes2,654 | ObservationsN/A | Result date |
| Arenasearch | Categoryoverall | Rank03 | Rating / score1,242.05 | Votes89,873 | ObservationsN/A | Result date |
| Arenasearch style control | Categoryoverall | Rank03 | Rating / score1,223.60 | Votes89,873 | ObservationsN/A | Result date |
| Arenatext factuality | Categorychinese | Rank03 | Rating / score1,541.39 | Votes2,950 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryexpert | Rank03 | Rating / score1,532.13 | Votes5,545 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry legal and government | Rank03 | Rating / score1,517.90 | Votes4,168 | ObservationsN/A | Result date |
| Arenatext factuality | Categorykorean | Rank03 | Rating / score1,462.98 | Votes629 | ObservationsN/A | Result date |
| Arenatext factuality | Categorypolish | Rank03 | Rating / score1,479.57 | Votes630 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryhard prompts | Rank04 | Rating / score1,513.33 | Votes41,567 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry business and management and financial operations | Rank04 | Rating / score1,499.48 | Votes11,330 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry mathematical | Rank04 | Rating / score1,489.22 | Votes2,921 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry software and it services | Rank04 | Rating / score1,522.39 | Votes24,912 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryinstruction following | Rank04 | Rating / score1,491.46 | Votes21,134 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank04 | Rating / score1,505.81 | Votes807 | ObservationsN/A | Result date |
| Arenadocument | Categoryoverall | Rank05 | Rating / score1,484.70 | Votes16,816 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank05 | Rating / score1,499.63 | Votes807 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank05 | Rating / score1,463.68 | Votes1,096 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryfrench | Rank05 | Rating / score1,516.37 | Votes1,798 | ObservationsN/A | Result date |
| Arenatext factuality | Categorynon english | Rank05 | Rating / score1,478.97 | Votes34,548 | ObservationsN/A | Result date |
| Arenaagent steerability | Categoryoverall | Rank06 | Rating / score0.09 | VotesN/A | Observations43.5K | Result date |
| Arenatext | Categorychinese | Rank06 | Rating / score1,537.35 | Votes3,544 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry life and physical and social science | Rank06 | Rating / score1,508.41 | Votes9,390 | ObservationsN/A | Result date |
| Arenatext factuality | Categorylonger query | Rank06 | Rating / score1,500.67 | Votes28,213 | ObservationsN/A | Result date |
| Arenatext factuality | Categorymath | Rank06 | Rating / score1,494.66 | Votes2,704 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryspanish | Rank06 | Rating / score1,477.96 | Votes1,360 | ObservationsN/A | Result date |
| Arenavision | Categorydiagram | Rank06 | Rating / score1,326.50 | Votes5,793 | ObservationsN/A | Result date |
GPT-5.5 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $5 input, $30 output per 1M
- Official provider
- OpenAI
- Lowest third-party
- From $0.1875 input, $1.13 output per 1M via UnoRouter
- Tracked offerings
- 47
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderKenari | Provider model IDgpt-5-5 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1.1M | Updated |
| ProviderUnoRouter | Provider model IDgpt-5.5:free | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1.1M | Updated |
| ProviderUnoRouter | Provider model IDgpt-5.5 | Regionglobal | Input / 1M$0.1875 | Output / 1M$1.13 | Context1.1M | Updated |
| ProviderXpersona | Provider model IDgpt-5.5 | Regionglobal | Input / 1M$1.5 | Output / 1M$12 | Context1.1M | Updated |
| ProviderFrogBot | Provider model IDgpt-5-5 | Regionglobal | Input / 1M$2.5 | Output / 1M$15 | Context272K | Updated |
| ProviderXpersona | Provider model IDxpersona-gpt-5.5 | Regionglobal | Input / 1M$3 | Output / 1M$18 | Context1M | Updated |
| ProviderPoe | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$4.55 | Output / 1M$27.27 | Context400K | Updated |
| ProviderHPC-AI | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderAI-ROUTER | Provider model IDgpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderOrcaRouter | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderNanoGPT | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderImpossibl | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderOpenRouter | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderDaoXE | Provider model IDgpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderCrossModel | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderDatabricks | Provider model IDdatabricks-gpt-5-5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderOpenAI | Provider model IDgpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderOpper | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderLLM Gateway | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderVivgrid | Provider model IDgpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderNEAR AI Cloud | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderMerge Gateway | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderCloudflare AI Gateway | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1M | Updated |
| ProviderFastRouter | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderZenMux | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderPerplexity Agent | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderGMI Cloud | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderSAP AI Core | Provider model IDgpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderFreeModel | Provider model IDgpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
| ProviderOfox | Provider model IDopenai/gpt-5.5 | Regionglobal | Input / 1M$5 | Output / 1M$30 | Context1.1M | Updated |
GPT-5.5 Runtime Performance
Provider-specific output speed and catalog latency for GPT-5.5. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderOpenAI | Output Speed134.94 tok/s | Catalog Latency4.56 s | Max Input1.1M | Max Output128K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
GPT-5.5 Specifications
Technical details for the model's default version.
- Version
- GPT-5.5
- Released
- Apr 23, 2026
- Knowledge cutoff
- Dec 1, 2025
- Parameters
- N/A
- Context window
- 1.1M
- Max output
- 128K
- Inputs
- image, text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
GPT-5.5 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGPT-5.5 | Released | LLMBoard81.73 | ParametersN/A | Context1.1M | Max output128K | Open weightsNo | LicenseProprietary |
Models similar to GPT-5.5
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
OP
GPT-5.6-Terra
OpenAI
OP
GPT-5.4
OpenAI
OP
GPT-5.5-Pro
OpenAI
OP
GPT-5.6-Luna
OpenAI
OP
GPT-5.6-Sol
OpenAI
OP
GPT-5.2-Pro
OpenAI
What is GPT-5.5?
Key information about GPT-5.5 and its available data.
5 is a large language model from OpenAI designed for agentic coding, computer use, knowledge work, and early scientific research. It supports a 1M-token context window in the API and a 400K-token context window in Codex.
Data as of 2026-09-08.
FAQ
Common questions about GPT-5.5.
When was GPT-5.5 released?
GPT-5.5's default version was released on Apr 23, 2026.
How much does GPT-5.5 cost?
GPT-5.5's official API price is $5 per million input tokens and $30 per million output tokens via OpenAI. The lowest tracked third-party offer starts at $0.1875 input and $1.13 output via UnoRouter.
Who created GPT-5.5?
GPT-5.5 was created by OpenAI.
What is the context window for GPT-5.5?
The default version has a 1.1M token context window.
Is GPT-5.5 open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer GPT-5.5?
47 provider offerings are linked to the default version.