Model profile
GPT-5.4 Pro
Evidence coverage
12 of 323 tracked benchmarks are published. 9 are verified and 3 provisional. 5 of 8 categories are measured.
- Published / tracked
- 12 / 323
- Verified
- 9
- Provisional
- 3
- Categories with evidence
- 5 / 8
Evidence by category
- Agentic1 benchmarkVerified
- Coding0 benchmarksNot measured
- Reasoning2 benchmarksMixed evidence
- Knowledge4 benchmarksMixed evidence
- Math4 benchmarksMixed evidence
- Multilingual0 benchmarksNot measured
- Multimodal1 benchmarkVerified
- Inst. Following0 benchmarksNot measured
GPT-5.4 Pro ranks #44 out of 200 models on the public leaderboard with an overall score of 60.89/100. It does not yet have enough sourced coverage for BenchLM's verified leaderboard. While not a frontier model, it offers specific advantages depending on the use case.
GPT-5.4 Pro is a proprietary model with a 1.05M token context window. It uses explicit chain-of-thought reasoning, which typically improves performance on math and complex reasoning tasks at the cost of higher latency and token usage.
GPT-5.4 Pro sits inside the GPT-5.4 family alongside GPT-5.4, GPT-5.4 mini, GPT-5.4 nano. This profile currently has 12 of 323 tracked benchmarks. BenchLM only exposes non-generated benchmark rows publicly, so missing categories stay blank until a sourced evaluation is available.
Its strongest category is Agentic (#17). This performance profile makes it particularly useful for coding agents, browser research, and computer-use workflows.
Peer position
Exact provisional scores and ranks for the closest listed peers. A score can appear before a model clears the evidence threshold for a rank, so equal scores can have different rank states.
Range 60.51–61.16
- GLM-4.7Z.AICompare#4261.16GLM-4.7 is #42 with a score of 61.16.
- Gemma 4 31BGoogleCompare#4361.08Gemma 4 31B is #43 with a score of 61.08.
- GPT-5.4 ProCurrent modelOpenAI#4460.89GPT-5.4 Pro is #44 with a score of 60.89.
- Qwen3.5-27BAlibabaCompare#4560.7Qwen3.5-27B is #45 with a score of 60.7.
- DeepSeek V4 ProDeepSeekCompare#4660.66DeepSeek V4 Pro is #46 with a score of 60.66.
- Qwen3.5-122B-A10BAlibabaCompare#4760.56Qwen3.5-122B-A10B is #47 with a score of 60.56.
- Grok 4.1 Fast (Reasoning)xAICompare#4860.51Grok 4.1 Fast (Reasoning) is #48 with a score of 60.51.
Category percentile
More
Relative position among models eligible for each sourced category. A higher percentile means a stronger position within that category's ranked cohort; 100 is highest.
- Agentic86%Eligible cohort rank #17 of 119Category score 58.2
Category evidence
Scores and ranks appear only where this model has published benchmark evidence. Categories without displayable source records remain not measured.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticRank #17 of 119Percentile 86thWeight 22%1 benchmarkVerified | 58.2 | #17 of 119 | 86th | 22% | 1 benchmark | Verified |
| CodingWeight 20%0 benchmarksNot measured | Not measured | Not ranked | Not available | 20% | 0 benchmarks | Not measured |
| ReasoningRank Not rankedWeight 17%2 benchmarksMixed sources | 92.7 | Not ranked | Not available | 17% | 2 benchmarks | Mixed sources |
| KnowledgeRank Not rankedWeight 12%4 benchmarksMixed sources | 93.5 | Not ranked | Not available | 12% | 4 benchmarks | Mixed sources |
| MathRank Not rankedWeight 5%4 benchmarksMixed sources | 70.8 | Not ranked | Not available | 5% | 4 benchmarks | Mixed sources |
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | Not ranked | Not available | 7% | 0 benchmarks | Not measured |
| MultimodalRank Not rankedWeight 12%1 benchmarkVerified | 87.9 | Not ranked | Not available | 12% | 1 benchmark | Verified |
| Inst. FollowingWeight 5%0 benchmarksNot measured | Not measured | Not ranked | Not available | 5% | 0 benchmarks | Not measured |
Chatbot Arena performance
Scroll horizontally to inspect confidence intervals and vote counts.
| View | Elo | Confidence interval | Votes |
|---|---|---|---|
| Text Overall | 1478 | ±4.0 | 58,116 |
| Coding | 1521 | ±6.2 | 15,756 |
| Math | 1495 | ±11.3 | 3,042 |
| Instruction Following | 1476 | ±5.8 | 19,104 |
| Creative Writing | 1444 | ±7.4 | 9,589 |
| Multi-turn | 1493 | ±7.0 | 10,720 |
| Hard Prompts | 1499 | ±4.8 | 37,471 |
| Hard Prompts (English) | 1496 | ±5.8 | 18,209 |
| Longer Query | 1484 | ±5.6 | 24,527 |
Benchmark Details
Rows below have a displayable published verification record. Each source link and provenance note remains in the page HTML while its category is closed. Source-unverified manual rows and generated rows stay hidden.
Agentic1 benchmark
Reasoning2 benchmarks
Abstraction and Reasoning Corpus for AGI v2
Critical Physics Tasks
Knowledge4 benchmarks
Humanity's Last Exam
Humanity's Last Exam without tools
Math4 benchmarks
FrontierMath v2 Tiers 1-3
FrontierMath v2 Tier 4
International Physics Olympiad 2025 (Theory)
FrontierMath legacy aggregate
Multimodal1 benchmark
Massive Multi-discipline Multimodal Understanding Pro
Frequently Asked Questions
How does GPT-5.4 Pro perform overall in AI benchmarks?
GPT-5.4 Pro has 12 published benchmark scores on BenchLM, but it does not yet have enough non-generated coverage to receive a global overall rank.
Is GPT-5.4 Pro good for knowledge and understanding?
GPT-5.4 Pro has visible benchmark coverage in knowledge and understanding, but BenchLM does not currently assign it a global category rank there.
Is GPT-5.4 Pro good for mathematics?
GPT-5.4 Pro has visible benchmark coverage in mathematics, but BenchLM does not currently assign it a global category rank there.
Is GPT-5.4 Pro good for reasoning and logic?
GPT-5.4 Pro has visible benchmark coverage in reasoning and logic, but BenchLM does not currently assign it a global category rank there.
Is GPT-5.4 Pro good for agentic tool use and computer tasks?
GPT-5.4 Pro ranks #17 out of 119 models in agentic tool use and computer tasks benchmarks with an average score of 58.2. There are stronger options in this category.
Is GPT-5.4 Pro good for multimodal and grounded tasks?
GPT-5.4 Pro has visible benchmark coverage in multimodal and grounded tasks, but BenchLM does not currently assign it a global category rank there.
Which sibling models are related to GPT-5.4 Pro?
GPT-5.4 Pro belongs to the GPT-5.4 family. Related variants on BenchLM include GPT-5.4, GPT-5.4 mini, GPT-5.4 nano.
Does GPT-5.4 Pro have full benchmark coverage on BenchLM?
Not yet. GPT-5.4 Pro currently has 12 published benchmark scores out of the 323 benchmarks BenchLM tracks. BenchLM only exposes non-generated public benchmark rows, so missing categories stay blank until a sourced evaluation is available.
What is the context window size of GPT-5.4 Pro?
GPT-5.4 Pro has a published context window of 1.05M, which determines how much text it can process in a single interaction.
Related Resources
Choose with this week’s evidence
Join 2,000+ readers for ranking moves, new releases, pricing changes, and the evidence behind them.
Free. One email per week.