GPT-5.4 Mini
OpenAI
53.1#121 of 427 overall
- Input / output
- US$ 0,75 / US$ 4,50 US$ per 1M tokens
- Context
- 400K tokens
- Release
- 03/17/2026 release date
- Weights
- Closed
- Speed
- 0 tok/s first token in 0.00 s
Family history
The 3 versions of this family, from first to newest, by release date. Click a point to open that version.
See the data as a table
| Model | Released | Score | Rank | Elo Arena |
|---|---|---|---|---|
| GPT-4.1 Mini | 04/14/2025 | 31.4 | #338 | 1340 |
| GPT-5 Mini | 08/07/2025 | 41.5 | #240 | 1373 |
| GPT-5.4 Mini | 03/17/2026 | 53.1 | #121 | 1412 |
Scores by source
Scores by sourcepercentile = % of ranked models it beats| Metric | Value | Percentile | Source |
|---|---|---|---|
| Arena text (Elo) weight 2 | 1412 | 51% | LMArena |
| Arena WebDev (Elo) weight 1 | 1397 | 31% | LMArena |
| Arena vision (Elo) weight 0.5 | 1245 | 47% | LMArena |
| Humanity's Last Exam weight 1 | 28.1 | 77% | Artificial Analysis |
| Terminal-Bench weight 1 | 2.0 | 62% | Artificial Analysis |
| GPQA Diamond | 87.5 | 85% | Artificial Analysis |
| Long-context reasoning | 77.0 | 76% | Artificial Analysis |
By category
| Category | Score | Rank | Metrics |
|---|---|---|---|
| Agentic | 55.8 | #49 | Terminal-Bench |
| Coding | 40.5 | #80 | Arena WebDev (Elo) |
| Reasoning | 70.5 | #77 | Humanity's Last Exam, GPQA Diamond |
| Research | 63.0 | #77 | Long-context reasoning |
| Design | 40.5 | #79 | Arena WebDev (Elo) |
| Chat | 50.6 | #96 | Arena text (Elo) |
| Vision | 48.4 | #57 | Arena vision (Elo) |
History
RankingIA score on each collection day (hover to see the rank).
Daily collection started on 10/30/2026. The chart appears from the second day the model is ranked.