OpenAI model product
GPT-5.4 mini is OpenAI's strongest mini model yet for coding, computer use, and subagents. It brings many of the strengths of GPT-5.4 to a faster, more efficient model designed for high-volume workloads, with strong coding, reasoning, multimodal understanding, and tool use.
Updated Aug 10, 2026. Default version: GPT-5.4 mini
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 mini.
| Graphwalks parents <128k | 0.7 | 7 | 18 | 64.7% | C | |
| Graphwalks BFS <128k | 0.8 | 8 | 22 | 66.7% | C | |
| Finance Agent v2 | 0.5 | 10 | 26 | 64.0% | B | |
| MRCR v2 (8-needle) | 0.3 | 11 | 21 | 50.0% | C | |
| OmniDocBench 1.5 | 0.9 | 11 | 16 | 33.3% | C | |
| Legal Agent Benchmark | 0.0 | 13 | 13 | 0.0% | B | |
| Tau2 Telecom | 0.9 | 13 | 35 | 64.7% | C | |
| OSWorld-Verified | 0.7 | 15 | 22 | 33.3% | C | |
| Terminal-Bench 2.0 | 0.6 | 20 | 49 | 60.4% | C | |
| Toolathlon | 0.4 | 23 | 31 | 26.7% | C | |
| MCP Atlas | 0.6 | 27 | 30 | 10.3% | C | |
| MMMU-Pro | 0.8 | 27 | 65 | 59.4% | C | |
| SWE-Bench Pro | 0.5 | 33 | 44 | 25.6% | C | |
| GPQA | 0.9 | 36 | 233 | 84.9% | C | |
| Humanity's Last Exam | 0.3 | 46 | 92 | 50.5% | C |
Preference and agent-evaluation signals from published Arena datasets.
The default version has no published Arena rows, or its source alias has not been resolved.
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-mini | global | $0.375 | $4 | 272K | |
| Vivgrid | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| LLM Gateway | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| OpenCode Zen | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| GitHub Copilot | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| 302.AI | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| Pioneer | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| Abacus | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| Azure Cognitive Services | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| Azure | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| AIHubMix | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| FreeModel | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| OpenAI | gpt-5.4-mini | global | $0.75 | $4.5 | 400K |
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 mini | 56.5 | N/A | 400K | 128K | No | Proprietary |
A concise description based on the published model registry.
GPT-5.4 mini is OpenAI's strongest mini model yet for coding, computer use, and subagents. It brings many of the strengths of GPT-5.4 to a faster, more efficient model designed for high-volume workloads, with strong coding, reasoning, multimodal understanding, and tool use.
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-mini.
GPT-5.4-mini's default version was released on Mar 17, 2026.
GPT-5.4-mini's official API price is $0.75 per million input tokens and $4.5 per million output tokens via OpenAI. The lowest tracked third-party offer starts at $0.375 input and $4 output via Xpersona.
GPT-5.4-mini is published under OpenAI in the model registry.
The default version has a 400K token context window.
No. The default version is not marked as having publicly available weights.
14 published provider offerings are linked to the default version.
Nearby ranked alternatives include Qwen3.5 27B, Qwen2.5 32B, GPT-OSS-120B.