OpenAI model product
4 mini is OpenAI's strongest mini model yet for coding, computer use, and subagents.
Updated Aug 12, 2026. Default version: GPT-5.4 mini
Technical details for the model's default version.
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for GPT-5.4 mini.
| Graphwalks BFS <128k | 76.3% | 04 | 11 | 70.0% | C | |
| Graphwalks parents <128k | 71.5% | 05 | 11 | 60.0% | C | |
| Finance Agent v2 | 45.4% | 10 | 26 | 64.0% | B | |
| MRCR v2 (8-needle) | 33.6% | 11 | 21 | 50.0% | C | |
| OmniDocBench 1.5 | 87.4% | 11 | 17 | 37.5% | C | |
| Legal Agent Benchmark | 0.0% | 13 | 13 | 0.0% | B | |
| Tau2 Telecom | 93.4% | 13 | 35 | 64.7% | C | |
| OSWorld-Verified | 72.1% | 15 | 23 | 36.4% | C | |
| Terminal-Bench 2.0 | 60.0% | 20 | 49 | 60.4% | C | |
| Toolathlon | 42.9% | 23 | 31 | 26.7% | C | |
| MMMU-Pro | 76.6% | 27 | 66 | 60.0% | C | |
| MCP Atlas | 57.7% | 28 | 31 | 10.0% | C | |
| SWE-Bench Pro | 54.4% | 33 | 45 | 27.3% | C | |
| GPQA | 88.0% | 36 | 234 | 85.0% | C | |
| Humanity's Last Exam | 28.2% | 46 | 93 | 51.1% | C |
Preference and agent-evaluation results for the default version.
| vision style control | overall | 37 | 1253.2 | 23,433 | N/A | |
| vision | overall | 52 | 1246.4 | 23,433 | N/A | |
| text factuality | overall | 54 | 1448.8 | 59,512 | N/A | |
| webdev | overall | 62 | 1397.6 | 11,270 | N/A | |
| text style control | overall | 65 | 1448.2 | 59,656 | N/A | |
| text | overall | 115 | 1412.1 | 59,656 | N/A |
Provider-specific output speed and catalog latency for GPT-5.4 mini. Runtime does not affect the capability score.
| OpenAI | 50.724 tok/s | 0.7 s | 400K | 128K |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Official vendor API pricing appears first, followed by individual provider offers.
| Kenari | gpt-5-4-mini | global | N/A | N/A | 400K | |
| Xpersona | gpt-5.4-mini | global | $0.375 | $4 | 272K | |
| Poe | openai/gpt-5.4-mini | global | $0.68 | $4 | 400K | |
| Impossibl | openai/gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| CrossModel | openai/gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| Vivgrid | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| NanoGPT | openai/gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| FastRouter | openai/gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| NEAR AI Cloud | openai/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 | |
| Neon | gpt-5-4-mini | global | $0.75 | $4.5 | 400K | |
| FrogBot | gpt-5-4-mini | global | $0.75 | $4.5 | 400K | |
| Pioneer | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| ZenMux | openai/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 | |
| Ofox | openai/gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| OrcaRouter | openai/gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| FreeModel | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| Kilo Gateway | openai/gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| Merge Gateway | openai/gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| Vercel AI Gateway | openai/gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| OpenRouter | openai/gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| OpenAI | gpt-5.4-mini | global | $0.75 | $4.5 | 400K | |
| Databricks | databricks-gpt-5-4-mini | global | $0.75 | $4.5 | 400K | |
| Venice AI | openai-gpt-54-mini | global | $0.9375 | $5.63 | 400K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Available versions of this model. The score column identifies the version used in the overall ranking.
| GPT-5.4 mini | 60.5 | N/A | 400K | 128K | No | Proprietary |
Key information about GPT-5.4-mini and its available data.
4 mini is OpenAI's strongest mini model yet for coding, computer use, and subagents. 4 to a faster, more efficient model designed for high-volume workloads, with strong coding, reasoning, multimodal understanding, and tool use.
Data as of 2026-08-11.
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 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 was created by OpenAI.
The default version has a 400K token context window.
No. The default version is not marked as having publicly available weights.
32 provider offerings are linked to the default version.
Nearby ranked alternatives include Kimi K2 Thinking, LongCat Flash Thinking, Qwen3.6 27B.