GPT-5.4 mini (High)
OpenAIProprietary
GPT-5.4 mini (High) is a vision-language model from OpenAI. It ranks 69 of 167 on LMArena’s overall vision board with 1245±6 from 25,734 votes. Its best category is Homework (#50 of 118). License: Proprietary.
Key figures
- Developer
- OpenAI
- License
- Proprietary
- First seen (our snapshots)
- 2026-10-26
No data: Captioning.
Who it suits
- Overall #69 of 167: upper half.
- Strongest category: Homework (#50 of 118).
- Proprietary; used through OpenAI’s apps or API.
Neighbours on the overall board
| # | Model | Score | Votes | License | Δ vs this model |
|---|---|---|---|---|---|
| 66 |  MiMo-V2.5 Xiaomi | 1247 | 29,381 | MIT | +2 |
| 67 |  Qwen3-VL-235B-A22B-Instruct Alibaba | 1247 | 12,724 | Apache 2.0 | +2 |
| 68 | GPT-5.2 (High) OpenAI | 1246 | 19,022 | Proprietary | +1 |
| 69 | GPT-5.4 mini (High) OpenAI | 1245 | 25,734 | Proprietary | |
| 70 | Qwen3.5-122B-A10B Alibaba | 1244 | 13,236 | Apache 2.0 | -1 |
| 71 | ChatGPT-4o Latest (20250326) OpenAI | 1243 | 23,400 | Proprietary | -2 |
| 72 | Qwen3.5-27B Alibaba | 1241 | 30,272 | Apache 2.0 | -4 |
Alternatives · More from OpenAI
| # | Model | Score | Votes | License |
|---|---|---|---|---|
| 18 | GPT-5.4 (High) OpenAI | 1303 | 30,747 | Proprietary |
| 21 | GPT-5.5 OpenAI | 1297 | 28,781 | Proprietary |
| 25 | GPT-5.5 (High) OpenAI | 1293 | 27,067 | Proprietary |
| 27 | GPT-5.4 OpenAI | 1292 | 22,804 | Proprietary |
| 29 | GPT-6.1 Sol (Max) OpenAI | 1288 | 2,330 | Proprietary |
Alternatives · Top open-weight models
| # | Model | Score | Votes | License |
|---|---|---|---|---|
| 24 |  GLM-5.3-Flash Z.ai | 1294 | 16,254 | MIT |
| 34 |  Kimi K2.6 Moonshot AI | 1282 | 16,368 | Modified MIT |
| 38 |  DeepSeek-V4.1-Flash-20260910 (Max) DeepSeek | 1278 | 4,100 | MIT |
| 39 |  Gemma 4 31B Google | 1276 | 40,219 | Apache 2.0 |
| 44 | Qwen3.8-27B Alibaba | 1269 | 7,614 | Apache 2.0 |
How to read these scores
- If two scores differ by less than their ± intervals, the models are statistically close.
- Each category board is computed independently; do not compare scores across categories.
- Models with few votes move more; ranks of new entries may change.
FAQ
Where does GPT-5.4 mini (High) rank among vision models?
Overall rank 69 (1245); OCR rank 68 (1253); Diagrams rank 63 (1266); Homework rank 50 (1285); Entity recognition rank 33 (1218); Creative writing rank 73 (1222); Humor rank 57 (1235); English rank 71 (1244); Chinese rank 59 (1286) (LMArena, data as of Oct 26, 2026).
How good is GPT-5.4 mini (High) at OCR?
GPT-5.4 mini (High) ranks 68 of 123 on LMArena’s OCR board with 1253 (18,272 votes).
Is GPT-5.4 mini (High) better with Chinese or English prompts?
Chinese board: #59 of 128 (1286); English board: #71 of 167 (1244). The boards are computed separately, so compare ranks rather than scores.
Is GPT-5.4 mini (High) open-weight?
No. LMArena lists it as Proprietary; it is available through OpenAI’s apps or API.
How far is GPT-5.4 mini (High) from #1?
#1 overall is Claude Fable 5 (High) (1325); GPT-5.4 mini (High) has 1245, 80 points behind.
Sources
LMArena data published Oct 19, 2026.
Synced here Oct 26, 2026.
Price, context length and release date have no reliable public data here and are not shown; all figures come from snapshots, blanks are –.