Model profile
GPT-5.4 nano
Evidence coverage
29 of 323 tracked benchmarks are published. 12 are verified and 17 provisional. 7 of 8 categories are measured.
- Published / tracked
- 29 / 323
- Verified
- 12
- Provisional
- 17
- Categories with evidence
- 7 / 8
Evidence by category
- Agentic9 benchmarksMixed evidence
- Coding3 benchmarksMixed evidence
- Reasoning2 benchmarksReported
- Knowledge9 benchmarksMixed evidence
- Math2 benchmarksVerified
- Multilingual0 benchmarksNot measured
- Multimodal3 benchmarksMixed evidence
- Inst. Following1 benchmarkReported
GPT-5.4 nano ranks #25 out of 200 models on the public leaderboard with an overall score of 66.79/100. It also ranks #22 out of 99 on the verified leaderboard. This places it in the mid-tier of AI models, with strengths in specific benchmark categories.
GPT-5.4 nano is a proprietary model with a 400K 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 nano sits inside the GPT-5.4 family alongside GPT-5.4, GPT-5.4 mini, GPT-5.4 Pro. BenchLM links it directly to GPT-5 nano as the earlier related model in that lineage. This profile currently has 29 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 Knowledge (#43), while its weakest is Coding (#97). This performance profile makes it particularly effective for knowledge-intensive tasks like research, analysis, and factual Q&A.
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 66.27–67.22
- Qwen3.7 PlusAlibabaCompare#2167.22Qwen3.7 Plus is #21 with a score of 67.22.
- GPT-5.6 LunaOpenAICompare#2267.17GPT-5.6 Luna is #22 with a score of 67.17.
- GPT-5.2 ProOpenAICompare#2367.01GPT-5.2 Pro is #23 with a score of 67.01.
- GLM-5-TurboZ.AICompare#2466.89GLM-5-Turbo is #24 with a score of 66.89.
- GPT-5.4 nanoCurrent modelOpenAI#2566.79GPT-5.4 nano is #25 with a score of 66.79.
- GPT-5.3 CodexOpenAICompare#2666.69GPT-5.3 Codex is #26 with a score of 66.69.
- Claude Opus 4.7 (Adaptive)AnthropicCompare#2766.27Claude Opus 4.7 (Adaptive) is #27 with a score of 66.27.
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.
- Knowledge18%Eligible cohort rank #43 of 52Category score 57.2
- Agentic25%Eligible cohort rank #89 of 119Category score 41.6
- Coding21%Eligible cohort rank #97 of 122Category score 43.6
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 #89 of 119Percentile 25thWeight 22%9 benchmarksMixed sources | 41.6 | #89 of 119 | 25th | 22% | 9 benchmarks | Mixed sources |
| CodingRank #97 of 122Percentile 21stWeight 20%3 benchmarksMixed sources | 43.6 | #97 of 122 | 21st | 20% | 3 benchmarks | Mixed sources |
| ReasoningRank Not rankedWeight 17%2 benchmarksReported | 38.7 | Not ranked | Not available | 17% | 2 benchmarks | Reported |
| KnowledgeRank #43 of 52Percentile 18thWeight 12%9 benchmarksMixed sources | 57.2 | #43 of 52 | 18th | 12% | 9 benchmarks | Mixed sources |
| MathRank Not rankedWeight 5%2 benchmarksVerified | 44.3 | Not ranked | Not available | 5% | 2 benchmarks | Verified |
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | Not ranked | Not available | 7% | 0 benchmarks | Not measured |
| MultimodalRank Not rankedWeight 12%3 benchmarksMixed sources | 15.4 | Not ranked | Not available | 12% | 3 benchmarks | Mixed sources |
| Inst. FollowingWeight 5%1 benchmarkReported | Score pending | Not ranked | Not available | 5% | 1 benchmark | Reported |
Chatbot Arena performance
Scroll horizontally to inspect confidence intervals and vote counts.
| View | Elo | Confidence interval | Votes |
|---|---|---|---|
| Text Overall | 1466 | ±8.2 | 5,618 |
| Coding | 1510 | ±16.1 | 1,321 |
| Math | 1472 | ±31.2 | 340 |
| Instruction Following | 1471 | ±14.8 | 1,591 |
| Creative Writing | 1421 | ±21.7 | 734 |
| Multi-turn | 1482 | ±18.3 | 1,088 |
| Hard Prompts | 1488 | ±10.5 | 3,230 |
| Hard Prompts (English) | 1487 | ±15.1 | 1,511 |
| Longer Query | 1483 | ±14.8 | 1,596 |
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.
Agentic9 benchmarks
τ²-Bench Tool-Agent-User Evaluation
Artificial Analysis Agentic Index
GDPval-AA normalized
Coding3 benchmarks
Vibe Code Bench v1.1
Artificial Analysis Coding Index
Artificial Analysis SciCode
Reasoning2 benchmarks
Artificial Analysis Long Context Reasoning
Critical Physics Tasks
Knowledge9 benchmarks
Humanity's Last Exam
Graduate-Level Google-Proof Q&A
Humanity's Last Exam without tools
Artificial Analysis GPQA Diamond
Artificial Analysis Humanity's Last Exam
Artificial Analysis Omniscience Index
Artificial Analysis Omniscience Accuracy
Artificial Analysis Omniscience Hallucination Rate
Math2 benchmarks
FrontierMath v2 Tiers 1-3
FrontierMath v2 Tier 4
Multimodal3 benchmarks
Massive Multi-discipline Multimodal Understanding Pro
MMMU-Pro with Python
Artificial Analysis MMMU-Pro
Inst. Following1 benchmark
Artificial Analysis IFBench
Frequently Asked Questions
How does GPT-5.4 nano perform overall in AI benchmarks?
GPT-5.4 nano currently ranks #25 out of 200 models on BenchLM's provisional leaderboard with an overall score of 66.79. It also ranks #22 out of 99 on the verified leaderboard. It is created by OpenAI. Its published context window is 400K.
Is GPT-5.4 nano good for knowledge and understanding?
GPT-5.4 nano ranks #43 out of 52 models in knowledge and understanding benchmarks with an average score of 57.2. There are stronger options in this category.
Is GPT-5.4 nano good for coding and programming?
GPT-5.4 nano ranks #97 out of 122 models in coding and programming benchmarks with an average score of 43.6. There are stronger options in this category.
Is GPT-5.4 nano good for mathematics?
GPT-5.4 nano has visible benchmark coverage in mathematics, but BenchLM does not currently assign it a global category rank there.
Is GPT-5.4 nano good for reasoning and logic?
GPT-5.4 nano has visible benchmark coverage in reasoning and logic, but BenchLM does not currently assign it a global category rank there.
Is GPT-5.4 nano good for agentic tool use and computer tasks?
GPT-5.4 nano ranks #89 out of 119 models in agentic tool use and computer tasks benchmarks with an average score of 41.6. There are stronger options in this category.
Is GPT-5.4 nano good for multimodal and grounded tasks?
GPT-5.4 nano has visible benchmark coverage in multimodal and grounded tasks, but BenchLM does not currently assign it a global category rank there.
Is GPT-5.4 nano good for instruction following?
GPT-5.4 nano has visible benchmark coverage in instruction following, but BenchLM does not currently assign it a global category rank there.
Which sibling models are related to GPT-5.4 nano?
GPT-5.4 nano belongs to the GPT-5.4 family. Related variants on BenchLM include GPT-5.4, GPT-5.4 mini, GPT-5.4 Pro.
Does GPT-5.4 nano have full benchmark coverage on BenchLM?
Not yet. GPT-5.4 nano currently has 29 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 nano?
GPT-5.4 nano has a published context window of 400K, 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.