OpenAI model product
GPT-5.4 is OpenAI's most capable and efficient frontier model for professional work. It combines industry-leading coding capabilities with native computer-use, up to 1M tokens of context, full-resolution vision processing, tool search for large tool ecosystems, and improved reasoning across spreadsheets, presentations, and documents. It is the most token-efficient reasoning model in the GPT-5 series.
Updated Aug 10, 2026. Default version: GPT-5.4
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-5.4.
| ARC-AGI | 0.9 | 2 | 7 | 83.3% | C | |
| Graphwalks BFS <128k | 0.9 | 2 | 22 | 95.2% | C | |
| Graphwalks parents <128k | 0.9 | 2 | 18 | 94.1% | C | |
| ARC-AGI v2 | 0.7 | 3 | 16 | 86.7% | C | |
| LiveBench | 0.8 | 3 | 38 | 94.6% | B | |
| Tau2 Telecom | 1.0 | 3 | 35 | 94.1% | C | |
| FrontierMath | 0.5 | 4 | 17 | 81.3% | C | |
| Finance Agent | 0.6 | 6 | 8 | 28.6% | C | |
| Terminal-Bench 2.0 | 0.8 | 6 | 49 | 89.6% | C | |
| OmniDocBench 1.5 | 0.9 | 7 | 16 | 60.0% | C | |
| Toolathlon | 0.5 | 8 | 31 | 76.7% | C | |
| FrontierSWE | 0.5 | 9 | 15 | 42.9% | B | |
| Legal Agent Benchmark | 0.0 | 9 | 13 | 33.3% | B | |
| MMMU-Pro | 0.8 | 9 | 65 | 87.5% | C | |
| GPQA | 0.9 | 11 | 233 | 95.7% | C | |
| OSWorld-Verified | 0.8 | 11 | 22 | 52.4% | C | |
| DeepSWE 1.1 | 0.5 | 13 | 19 | 33.3% | B | |
| Graphwalks parents >128k | 0.3 | 14 | 18 | 23.5% | C | |
| BrowseComp | 0.8 | 19 | 58 | 68.4% | C | |
| Graphwalks BFS >128k | 0.2 | 19 | 22 | 14.3% | C | |
| SWE-Bench Pro | 0.6 | 20 | 44 | 55.8% | C | |
| MCP Atlas | 0.7 | 21 | 30 | 31.0% | C | |
| Humanity's Last Exam | 0.4 | 35 | 92 | 62.6% | C |
Preference and agent-evaluation signals from published Arena datasets.
| text factuality | korean | 5 | 1450.5 | 565 | N/A | |
| text factuality | french | 8 | 1506.9 | 1,746 | N/A | |
| text factuality | japanese | 9 | 1463.5 | 422 | N/A | |
| text factuality | hard prompts | 10 | 1502.8 | 41,194 | N/A | |
| text factuality | industry business and management and financial operations | 10 | 1485.5 | 11,083 | N/A | |
| text factuality | polish | 10 | 1460.9 | 763 | N/A | |
| text factuality | industry legal and government | 11 | 1501.5 | 3,991 | N/A | |
| text factuality | spanish | 11 | 1468.1 | 1,419 | N/A | |
| text factuality | chinese | 12 | 1526.8 | 2,821 | N/A | |
| vision | diagram | 12 | 1319.0 | 5,615 | N/A | |
| vision style control | diagram | 13 | 1318.0 | 5,615 | N/A | |
| text factuality | overall | 14 | 1475.9 | 62,953 | N/A | |
| text factuality | expert | 14 | 1509.4 | 4,801 | N/A | |
| text factuality | math | 14 | 1476.3 | 2,590 | N/A | |
| text factuality | multi turn | 14 | 1488.6 | 9,989 | N/A | |
| document | overall | 15 | 1470.1 | 29,809 | N/A | |
| text factuality | coding | 15 | 1526.1 | 16,194 | N/A | |
| text factuality | german | 15 | 1424.3 | 496 | N/A | |
| text factuality | industry software and it services | 15 | 1511.6 | 24,234 | N/A | |
| text factuality | non english | 15 | 1466.8 | 33,680 | N/A | |
| document style control | overall | 16 | 1469.3 | 29,809 | N/A | |
| text factuality | industry mathematical | 16 | 1475.8 | 2,632 | N/A | |
| text | korean | 18 | 1435.1 | 1,133 | N/A | |
| text factuality | english | 18 | 1480.1 | 29,164 | N/A | |
| vision | homework | 18 | 1312.2 | 2,847 | N/A | |
| text factuality | exclude ties | 19 | 1477.1 | 48,367 | N/A | |
| vision style control | ocr | 19 | 1293.7 | 14,870 | N/A | |
| text factuality | hard prompts english | 20 | 1503.9 | 18,700 | N/A | |
| text factuality | longer query | 20 | 1486.8 | 25,428 | N/A | |
| vision | overall | 20 | 1293.0 | 20,516 | N/A | |
| vision | ocr | 20 | 1301.5 | 14,870 | N/A | |
| vision style control | overall | 20 | 1280.5 | 20,516 | N/A | |
| vision style control | english | 20 | 1276.7 | 8,737 | N/A | |
| text | japanese | 21 | 1454.0 | 609 | N/A | |
| text factuality | industry writing and literature and language | 21 | 1460.6 | 13,568 | N/A | |
| text factuality | instruction following | 21 | 1474.4 | 19,951 | N/A | |
| vision style control | homework | 21 | 1307.0 | 2,847 | N/A | |
| text factuality | industry life and physical and social science | 22 | 1490.8 | 8,523 | N/A | |
| vision | chinese | 22 | 1341.3 | 1,059 | N/A | |
| vision | english | 22 | 1288.7 | 8,737 | N/A | |
| text factuality | industry medicine and healthcare | 24 | 1493.4 | 3,634 | N/A | |
| text factuality | russian | 24 | 1475.2 | 5,427 | N/A | |
| text style control | industry business and management and financial operations | 26 | 1473.7 | 12,871 | N/A | |
| text | industry business and management and financial operations | 27 | 1456.1 | 12,871 | N/A | |
| text factuality | industry entertainment and sports and media | 27 | 1441.4 | 11,971 | N/A | |
| text style control | industry writing and literature and language | 27 | 1459.4 | 15,139 | N/A | |
| text style control | japanese | 27 | 1450.3 | 609 | N/A | |
| vision | creative writing vision | 29 | 1268.5 | 1,149 | N/A | |
| webdev | image to webdev | 29 | 1448.7 | 1,802 | N/A | |
| text | chinese | 30 | 1509.9 | 3,514 | N/A | |
| text | non english | 30 | 1448.3 | 33,780 | N/A | |
| vision style control | chinese | 30 | 1299.5 | 1,059 | N/A | |
| text | industry writing and literature and language | 31 | 1448.9 | 15,139 | N/A | |
| text | polish | 31 | 1454.3 | 1,238 | N/A | |
| vision style control | creative writing vision | 31 | 1253.3 | 1,149 | N/A | |
| text | expert | 32 | 1480.9 | 5,888 | N/A | |
| text style control | multi turn | 32 | 1480.2 | 11,861 | N/A | |
| text style control | polish | 32 | 1471.4 | 1,238 | N/A | |
| text | french | 33 | 1470.8 | 2,274 | N/A | |
| text | russian | 33 | 1458.4 | 6,743 | N/A | |
| text | overall | 34 | 1451.7 | 63,097 | N/A | |
| text | hard prompts | 34 | 1468.2 | 41,311 | N/A | |
| text | instruction following | 34 | 1454.2 | 21,300 | N/A | |
| text style control | coding | 34 | 1513.8 | 17,549 | N/A | |
| text style control | longer query | 34 | 1477.6 | 26,792 | N/A | |
| text style control | non english | 34 | 1457.4 | 33,780 | N/A | |
| vision style control | humor | 34 | 1239.6 | 702 | N/A | |
| text | industry legal and government | 35 | 1462.9 | 4,965 | N/A | |
| text | longer query | 35 | 1463.0 | 26,792 | N/A | |
| text | multi turn | 35 | 1462.2 | 11,861 | N/A | |
| text style control | russian | 35 | 1470.7 | 6,743 | N/A | |
| vision | humor | 35 | 1254.5 | 702 | N/A | |
| text | exclude ties | 36 | 1450.2 | 48,482 | N/A | |
| text style control | french | 36 | 1484.8 | 2,274 | N/A | |
| text style control | industry legal and government | 36 | 1475.6 | 4,965 | N/A | |
| text style control | spanish | 36 | 1462.5 | 1,878 | N/A | |
| text factuality | creative writing | 37 | 1441.1 | 8,475 | N/A | |
| text style control | instruction following | 37 | 1461.2 | 21,300 | N/A | |
| text style control | korean | 38 | 1423.3 | 1,133 | N/A | |
| text | industry mathematical | 39 | 1466.5 | 3,349 | N/A | |
| text | industry software and it services | 39 | 1474.1 | 25,052 | N/A | |
| text style control | expert | 39 | 1492.0 | 5,888 | N/A | |
| text style control | hard prompts | 39 | 1487.5 | 41,311 | N/A | |
| text | coding | 40 | 1479.9 | 17,549 | N/A | |
| text style control | industry mathematical | 40 | 1473.0 | 3,349 | N/A | |
| text | hard prompts english | 41 | 1463.0 | 20,026 | N/A | |
| text | industry life and physical and social science | 41 | 1461.5 | 10,157 | N/A | |
| text | spanish | 41 | 1453.0 | 1,878 | N/A | |
| text style control | overall | 42 | 1465.1 | 63,097 | N/A | |
| text style control | exclude ties | 42 | 1471.4 | 48,482 | N/A | |
| text style control | industry software and it services | 43 | 1498.8 | 25,052 | N/A | |
| text | industry entertainment and sports and media | 44 | 1424.3 | 13,399 | N/A | |
| text | math | 44 | 1457.3 | 3,334 | N/A | |
| text style control | english | 45 | 1468.8 | 29,315 | N/A | |
| text style control | hard prompts english | 45 | 1486.8 | 20,026 | N/A | |
| text style control | industry entertainment and sports and media | 48 | 1433.5 | 13,399 | N/A | |
| text style control | math | 48 | 1460.8 | 3,334 | N/A | |
| text | creative writing | 49 | 1426.7 | 10,176 | N/A | |
| text style control | creative writing | 50 | 1434.8 | 10,176 | N/A | |
| text style control | industry life and physical and social science | 50 | 1478.4 | 10,157 | 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.
| Xpersona | gpt-5.4 | global | $0.75 | $6 | 1.1M | |
| UnoRouter | gpt-5.4 | global | $1.8 | $10.8 | 1.1M | |
| Vivgrid | gpt-5.4 | global | $2.5 | $15 | 400K | |
| DaoXE | gpt-5.4 | global | $2.5 | $15 | 1.1M | |
| LLM Gateway | gpt-5.4 | global | $2.5 | $15 | 1.1M | |
| OpenCode Zen | gpt-5.4 | global | $2.5 | $15 | 1.1M | |
| GitHub Copilot | gpt-5.4 | global | $2.5 | $15 | 1.1M | |
| 302.AI | gpt-5.4 | global | $2.5 | $15 | 1.1M | |
| SAP AI Core | gpt-5.4 | global | $2.5 | $15 | 1.1M | |
| Pioneer | gpt-5.4 | global | $2.5 | $15 | 1.1M | |
| AI-ROUTER | gpt-5.4 | global | $2.5 | $15 | 1.1M | |
| Abacus | gpt-5.4 | global | $2.5 | $15 | 400K | |
| Azure Cognitive Services | gpt-5.4 | global | $2.5 | $15 | 1.1M | |
| Azure | gpt-5.4 | global | $2.5 | $15 | 1.1M | |
| AIHubMix | gpt-5.4 | global | $2.5 | $15 | 1.1M | |
| FreeModel | gpt-5.4 | global | $2.5 | $15 | 1.1M | |
| OpenAI | gpt-5.4 | global | $2.5 | $15 | 1.1M | |
| Cortecs | gpt-5.4 | global | $2.9 | $15.45 | 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-5.4 | 75.6 | N/A | 1.1M | 128K | No | Proprietary |
A concise description based on the published model registry.
GPT-5.4 is OpenAI's most capable and efficient frontier model for professional work. It combines industry-leading coding capabilities with native computer-use, up to 1M tokens of context, full-resolution vision processing, tool search for large tool ecosystems, and improved reasoning across spreadsheets, presentations, and documents. It is the most token-efficient reasoning model in the GPT-5 series.
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-5.4.
GPT-5.4's default version was released on Mar 5, 2026.
GPT-5.4's official API price is $2.5 per million input tokens and $15 per million output tokens via OpenAI. The lowest tracked third-party offer starts at $0.75 input and $6 output via Xpersona.
GPT-5.4 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.
19 published provider offerings are linked to the default version.
Nearby ranked alternatives include Gemini 2.5 Pro, GPT-5.6-Luna, GPT-5.5.