OpenAI model product
GPT-5.5 is OpenAI's smartest model yet, designed for real work across agentic coding, computer use, knowledge work, and early scientific research. It matches GPT-5.4 per-token latency in real-world serving while reaching a much higher level of intelligence and using significantly fewer tokens to complete the same tasks. GPT-5.5 supports a 1M-token context window in the API and a 400K-token context window in Codex, with state-of-the-art results on Terminal-Bench 2.0, OSWorld-Verified, GDPval, FrontierMath, and CyberGym.
Updated Aug 10, 2026. Default version: GPT-5.5
Structured fields from the published default version.
This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.
Published benchmark records for the scored version GPT-5.5.
| ARC-AGI | 0.9 | 1 | 7 | 100.0% | C | |
| ARC-AGI v2 | 0.8 | 1 | 16 | 100.0% | C | |
| BixBench | 0.8 | 1 | 1 | 100.0% | C | |
| GDPval-MM | 0.8 | 1 | 3 | 100.0% | C | |
| Terminal-Bench 2.0 | 0.8 | 1 | 49 | 100.0% | C | |
| GeneBench | 0.3 | 2 | 2 | 0.0% | C | |
| LiveBench | 0.8 | 2 | 38 | 97.3% | B | |
| MMMU-Pro | 0.8 | 2 | 65 | 98.4% | C | |
| CyberGym | 0.8 | 3 | 11 | 80.0% | C | |
| Finance Agent | 0.6 | 4 | 8 | 57.1% | C | |
| MRCR v2 (8-needle) | 0.7 | 4 | 21 | 85.0% | C | |
| Finance Agent v2 | 0.5 | 5 | 26 | 84.0% | B | |
| FrontierCode 1.1 | 0.4 | 5 | 15 | 71.4% | B | |
| DeepSWE 1.1 | 0.7 | 6 | 19 | 72.2% | B | |
| FrontierSWE | 0.7 | 6 | 15 | 64.3% | B | |
| GPQA | 0.9 | 6 | 233 | 97.8% | C | |
| OfficeQA Pro | 0.5 | 6 | 7 | 16.7% | C | |
| Tau2 Telecom | 1.0 | 6 | 35 | 85.3% | C | |
| FrontierMath | 0.4 | 7 | 17 | 62.5% | C | |
| Legal Agent Benchmark | 0.0 | 7 | 13 | 50.0% | B | |
| Toolathlon | 0.6 | 7 | 31 | 80.0% | C | |
| OSWorld-Verified | 0.8 | 8 | 22 | 66.7% | C | |
| Graphwalks parents >128k | 0.6 | 9 | 18 | 52.9% | C | |
| BrowseComp | 0.8 | 12 | 58 | 80.7% | C | |
| MCP Atlas | 0.8 | 12 | 30 | 62.1% | C | |
| Graphwalks BFS >128k | 0.5 | 16 | 22 | 28.6% | C | |
| Humanity's Last Exam | 0.5 | 16 | 92 | 83.5% | C | |
| SWE-Bench Pro | 0.6 | 16 | 44 | 65.1% | C |
Preference and agent-evaluation signals from published Arena datasets.
| text factuality | japanese | 2 | 1497.2 | 389 | N/A | |
| text factuality | german | 4 | 1460.6 | 390 | N/A | |
| text factuality | korean | 4 | 1452.9 | 471 | N/A | |
| text factuality | chinese | 5 | 1538.7 | 2,268 | N/A | |
| text factuality | french | 5 | 1516.5 | 1,578 | N/A | |
| text factuality | industry business and management and financial operations | 5 | 1496.8 | 9,487 | N/A | |
| text factuality | math | 6 | 1491.3 | 2,125 | N/A | |
| text style control | japanese | 6 | 1497.4 | 513 | N/A | |
| text | chinese | 8 | 1537.8 | 2,751 | N/A | |
| text | japanese | 8 | 1490.5 | 513 | N/A | |
| text factuality | expert | 8 | 1523.3 | 4,480 | N/A | |
| text factuality | instruction following | 8 | 1487.8 | 17,424 | N/A | |
| vision | diagram | 8 | 1327.8 | 5,177 | N/A | |
| vision style control | diagram | 8 | 1326.4 | 5,177 | N/A | |
| vision style control | homework | 8 | 1329.9 | 2,538 | N/A | |
| document | overall | 9 | 1480.3 | 17,286 | N/A | |
| text factuality | english | 9 | 1486.9 | 24,795 | N/A | |
| text factuality | hard prompts | 9 | 1508.7 | 35,252 | N/A | |
| text factuality | hard prompts english | 9 | 1512.1 | 16,410 | N/A | |
| text factuality | industry life and physical and social science | 9 | 1505.6 | 7,667 | N/A | |
| text factuality | overall | 10 | 1481.8 | 53,330 | N/A | |
| text factuality | longer query | 10 | 1495.8 | 23,332 | N/A | |
| text style control | math | 10 | 1499.2 | 2,722 | N/A | |
| text factuality | industry mathematical | 11 | 1483.0 | 2,295 | N/A | |
| text factuality | industry software and it services | 11 | 1515.8 | 21,074 | N/A | |
| text factuality | russian | 11 | 1491.5 | 4,461 | N/A | |
| vision style control | english | 11 | 1283.5 | 7,886 | N/A | |
| text factuality | industry legal and government | 12 | 1500.9 | 3,504 | N/A | |
| text style control | chinese | 12 | 1531.7 | 2,751 | N/A | |
| vision style control | overall | 12 | 1286.0 | 18,722 | N/A | |
| text factuality | exclude ties | 13 | 1486.9 | 40,759 | N/A | |
| text factuality | industry writing and literature and language | 13 | 1469.6 | 12,640 | N/A | |
| text factuality | non english | 13 | 1471.6 | 28,523 | N/A | |
| text factuality | spanish | 13 | 1461.6 | 1,174 | N/A | |
| text factuality | industry entertainment and sports and media | 14 | 1455.9 | 11,442 | N/A | |
| text factuality | polish | 14 | 1452.0 | 586 | N/A | |
| vision | homework | 14 | 1326.4 | 2,538 | N/A | |
| vision style control | humor | 14 | 1276.6 | 642 | N/A | |
| document style control | overall | 15 | 1475.5 | 17,286 | N/A | |
| text | expert | 15 | 1501.8 | 5,361 | N/A | |
| text | industry business and management and financial operations | 15 | 1473.4 | 10,608 | N/A | |
| text | instruction following | 15 | 1473.1 | 18,057 | N/A | |
| text factuality | multi turn | 15 | 1488.1 | 8,177 | N/A | |
| vision | overall | 15 | 1296.3 | 18,722 | N/A | |
| vision | english | 15 | 1294.1 | 7,886 | N/A | |
| vision | humor | 15 | 1286.5 | 642 | N/A | |
| vision style control | ocr | 15 | 1298.6 | 13,517 | N/A | |
| text | french | 16 | 1489.0 | 2,005 | N/A | |
| text | math | 16 | 1484.5 | 2,722 | N/A | |
| text factuality | industry medicine and healthcare | 16 | 1498.5 | 3,187 | N/A | |
| text style control | german | 16 | 1481.6 | 813 | N/A | |
| text style control | industry business and management and financial operations | 16 | 1485.7 | 10,608 | N/A | |
| text style control | industry mathematical | 16 | 1497.9 | 2,912 | N/A | |
| vision | ocr | 16 | 1306.0 | 13,517 | N/A | |
| text | german | 17 | 1470.1 | 813 | N/A | |
| text | korean | 17 | 1441.5 | 889 | N/A | |
| text style control | expert | 17 | 1510.5 | 5,361 | N/A | |
| text factuality | coding | 18 | 1525.1 | 14,523 | N/A | |
| text style control | french | 18 | 1503.2 | 2,005 | N/A | |
| vision | chinese | 18 | 1349.4 | 1,077 | N/A | |
| text | industry legal and government | 19 | 1479.9 | 4,321 | N/A | |
| text | industry writing and literature and language | 19 | 1462.2 | 13,360 | N/A | |
| text | longer query | 19 | 1478.5 | 23,890 | N/A | |
| text factuality | creative writing | 19 | 1453.8 | 8,256 | N/A | |
| text style control | overall | 20 | 1476.7 | 53,407 | N/A | |
| text style control | industry writing and literature and language | 20 | 1466.6 | 13,360 | N/A | |
| text | hard prompts | 21 | 1482.9 | 35,299 | N/A | |
| text | industry mathematical | 21 | 1485.7 | 2,912 | N/A | |
| text style control | instruction following | 21 | 1473.9 | 18,057 | N/A | |
| vision | creative writing vision | 21 | 1282.7 | 1,106 | N/A | |
| text | industry life and physical and social science | 22 | 1481.5 | 8,694 | N/A | |
| text | russian | 22 | 1473.1 | 5,369 | N/A | |
| text style control | exclude ties | 22 | 1485.9 | 40,814 | N/A | |
| text style control | industry life and physical and social science | 22 | 1494.5 | 8,694 | N/A | |
| vision style control | creative writing vision | 22 | 1266.6 | 1,106 | N/A | |
| text | non english | 23 | 1459.9 | 28,567 | N/A | |
| text style control | english | 23 | 1481.0 | 24,839 | N/A | |
| text style control | non english | 23 | 1467.3 | 28,567 | N/A | |
| text | overall | 24 | 1465.8 | 53,407 | N/A | |
| text | exclude ties | 24 | 1469.0 | 40,814 | N/A | |
| text | industry entertainment and sports and media | 24 | 1442.3 | 12,179 | N/A | |
| text style control | industry legal and government | 24 | 1485.2 | 4,321 | N/A | |
| vision style control | chinese | 24 | 1315.6 | 1,077 | N/A | |
| text | creative writing | 25 | 1449.2 | 9,371 | N/A | |
| text style control | hard prompts | 25 | 1496.5 | 35,299 | N/A | |
| text style control | longer query | 25 | 1484.0 | 23,890 | N/A | |
| text style control | russian | 25 | 1479.8 | 5,369 | N/A | |
| text | hard prompts english | 26 | 1478.4 | 17,055 | N/A | |
| text | industry software and it services | 26 | 1485.6 | 21,389 | N/A | |
| text | english | 27 | 1466.3 | 24,839 | N/A | |
| text | multi turn | 28 | 1468.6 | 9,417 | N/A | |
| text style control | hard prompts english | 28 | 1495.8 | 17,055 | N/A | |
| text style control | multi turn | 29 | 1480.7 | 9,417 | N/A | |
| text | spanish | 30 | 1458.8 | 1,569 | N/A | |
| text style control | industry software and it services | 30 | 1505.4 | 21,389 | N/A | |
| text | polish | 32 | 1454.1 | 972 | N/A | |
| text style control | industry entertainment and sports and media | 32 | 1443.8 | 12,179 | N/A | |
| text style control | creative writing | 33 | 1447.0 | 9,371 | N/A | |
| text style control | industry medicine and healthcare | 33 | 1485.6 | 3,947 | N/A | |
| text style control | korean | 33 | 1427.2 | 889 | N/A |
Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.
| UnoRouter | gpt-5.5 | global | $0.1875 | $1.13 | 1.1M | |
| Xpersona | gpt-5.5 | global | $1.5 | $12 | 1.1M | |
| Vivgrid | gpt-5.5 | global | $5 | $30 | 1.1M | |
| DaoXE | gpt-5.5 | global | $5 | $30 | 1.1M | |
| LLM Gateway | gpt-5.5 | global | $5 | $30 | 1.1M | |
| OpenCode Zen | gpt-5.5 | global | $5 | $30 | 1.1M | |
| GitHub Copilot | gpt-5.5 | global | $5 | $30 | 1.1M | |
| SAP AI Core | gpt-5.5 | global | $5 | $30 | 1.1M | |
| Pioneer | gpt-5.5 | global | $5 | $30 | 1.1M | |
| AI-ROUTER | gpt-5.5 | global | $5 | $30 | 1.1M | |
| Abacus | gpt-5.5 | global | $5 | $30 | 1M | |
| Azure Cognitive Services | gpt-5.5 | global | $5 | $30 | 1.1M | |
| Azure | gpt-5.5 | global | $5 | $30 | 1.1M | |
| AIHubMix | gpt-5.5 | global | $5 | $30 | 1.1M | |
| FreeModel | gpt-5.5 | global | $5 | $30 | 1.1M | |
| OpenAI | gpt-5.5 | global | $5 | $30 | 1.1M |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.
| GPT-5.5 | 76.8 | N/A | 1.1M | 128K | No | Proprietary |
A concise description based on the published model registry.
GPT-5.5 is OpenAI's smartest model yet, designed for real work across agentic coding, computer use, knowledge work, and early scientific research. It matches GPT-5.4 per-token latency in real-world serving while reaching a much higher level of intelligence and using significantly fewer tokens to complete the same tasks. GPT-5.5 supports a 1M-token context window in the API and a 400K-token context window in Codex, with state-of-the-art results on Terminal-Bench 2.0, OSWorld-Verified, GDPval, FrontierMath, and CyberGym.
Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.
Data snapshot: 2026-08-07. Editorial model content is not available in the backend.
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest published LLMBoard score.
Common questions about GPT-5.5.
GPT-5.5's default version was released on Apr 23, 2026.
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.
GPT-5.5 is published under OpenAI in the model registry.
The default version has a 1.1M token context window.
No. The default version is not marked as having publicly available weights.
17 published provider offerings are linked to the default version.
Nearby ranked alternatives include Claude Opus 4.7, Gemini 2.5 Pro, GPT-5.6-Luna.