GPT-5.4 (High)

OpenAIProprietary

GPT-5.4 (High) is a vision-language model from OpenAI. It ranks 19 of 167 on LMArena’s overall vision board with 1301±6 from 32,045 votes. Its best category is Homework (#8 of 118). License: Proprietary.

Key figures

LMArena · Overall1301±6#19 / 167 · Votes 32,045
OCR1310±6#16 / 123 · Votes 22,696
Diagrams1325±8#12 / 123 · Votes 8,574
Homework1336±10#8 / 118 · Votes 4,223
Entity recognition1226±45#29 / 52 · Votes 163
Creative writing1285±13#27 / 106 · Votes 2,246
Humor1288±18#20 / 102 · Votes 1,219
English1299±8#16 / 167 · Votes 12,546
Chinese1354±16#22 / 135 · Votes 1,914
Developer
OpenAI
License
Proprietary
First seen (our snapshots)
2026-10-26

No data: Captioning.

Who it suits

  • Overall #19 of 167: upper half.
  • Strongest category: Homework (#8 of 118).
  • Proprietary; used through OpenAI’s apps or API.

Neighbours on the overall board

#ModelScoreVotesLicenseΔ vs this model
16Muse Spark 1.2 (xHigh) Meta13051,930Proprietary+4
17Gemini 3 Pro Google130513,547Proprietary+4
18Muse Spark 1.3 (xHigh) Meta13017,522Proprietary+0
19GPT-5.4 (High) OpenAI130132,045Proprietary
20Gemini 3.6 Flash (High) Google129710,352Proprietary-4
21Gemini 3.1 Pro Preview Google129648,242Proprietary-5
22GPT-5.5 OpenAI129630,058Proprietary-5

Alternatives · More from OpenAI

#ModelScoreVotesLicense
22GPT-5.5 OpenAI129630,058Proprietary
25GPT-5.5 (High) OpenAI129528,348Proprietary
27GPT-5.4 OpenAI129222,608Proprietary
30GPT-6.1 Sol (Max) OpenAI12863,987Proprietary
35GPT-5.6 Sol (xHigh) OpenAI128212,342Proprietary

Alternatives · Top open-weight models

#ModelScoreVotesLicense
24GLM-5.3-Flash Z.ai129531,530MIT
34Kimi K2.6 Moonshot AI128216,265Modified MIT
36Gemma 4 31B Google127941,787Apache 2.0
37DeepSeek-V4.1-Flash-20260910 (Max) DeepSeek12794,247MIT
42Qwen3.8-27B Alibaba12709,632Apache 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 (High) rank among vision models?

Overall rank 19 (1301); OCR rank 16 (1310); Diagrams rank 12 (1325); Homework rank 8 (1336); Entity recognition rank 29 (1226); Creative writing rank 27 (1285); Humor rank 20 (1288); English rank 16 (1299); Chinese rank 22 (1354) (LMArena, data as of Nov 14, 2026).

How good is GPT-5.4 (High) at OCR?

GPT-5.4 (High) ranks 16 of 123 on LMArena’s OCR board with 1310 (22,696 votes).

Is GPT-5.4 (High) better with Chinese or English prompts?

Chinese board: #22 of 135 (1354); English board: #16 of 167 (1299). The boards are computed separately, so compare ranks rather than scores.

Is GPT-5.4 (High) open-weight?

No. LMArena lists it as Proprietary; it is available through OpenAI’s apps or API.

How far is GPT-5.4 (High) from #1?

#1 overall is Claude Fable 5 (High) (1326); GPT-5.4 (High) has 1301, 25 points behind.

Sources

LMArena data published Nov 9, 2026.

Synced here Nov 14, 2026.

Price, context length and release date have no reliable public data here and are not shown; all figures come from snapshots, blanks are –.