OpenAI, proprietary

# GPT-5.4 nano

> GPT-5.4 nano by OpenAI, released March 2026. Ranked #125 of 354 with a Noometry Index of 41.9. API: $0.20 in / $1.25 out per M tokens. 400K context. Scores, sources and comparisons.
- Canonical page: https://noometry.com/models/gpt-5-4-nano
- Last updated: 2026-10-10
- Title: GPT-5.4 nano Benchmarks, Price & Rank (October 2026)

GPT-5.4 nano by OpenAI ranks 125th of 354 ranked models on the Noometry Index as of October 2026, with a score of 41.9. Its strongest category is multimodal, where it ranks 78th. API pricing starts at $0.20 per million input tokens and $1.25 per million output tokens, with a 400K-token context window.

Last verified October 10, 2026

## Specifications

- **Noometry rank:** #125 of 354
- **Index score:** 41.9
- **Evidence:** Confirmed 40 results
- **Provider:** [OpenAI](https://noometry.com/providers/openai)
- **Released:** March 17, 2026
- **Weights:** Proprietary
- **Reasoning:** Yes
- **Context window:** 400K
- **Max output:** 128K
- **Input price:** $0.20 / M
- **Output price:** $1.25 / M
- **Blended price:** $0.46 / M
- **Output speed:** 19 tokens/s [Kagi](https://help.kagi.com/kagi/ai/llm-benchmark.html)
- **Value:** #71 of 219
- **Knowledge cutoff:** August 2025
- **Input:** text, image

## Category scores

Each category score combines every public result we have in that category.

GPT-5.4 nano category scores

1.  Coding 43.6
2.  Reasoning 23.7
3.  Math 40.9
4.  Knowledge 41.9
5.  Multimodal 36.7
6.  Multilingual 48.6
7.  Instruction Following 71.9
8.  Long Context 41.6
9.  Writing & Preference 55.7
10.  020406080

GPT-5.4 nano category ranks
| Category | Score | Rank | Results |
| --- | --- | --- | --- |
| [Coding](https://noometry.com/best/coding) | 43.6 | #84 | 3 |
| [Reasoning](https://noometry.com/best/reasoning) | 23.7 | #173 | 9 |
| [Math](https://noometry.com/best/math) | 40.9 | #88 | 5 |
| [Knowledge](https://noometry.com/best/knowledge) | 41.9 | #103 | 4 |
| [Multimodal](https://noometry.com/best/multimodal) | 36.7 | #78 | 1 |
| [Multilingual](https://noometry.com/best/multilingual) | 48.6 | #140 | 1 |
| [Instruction Following](https://noometry.com/best/instruction-following) | 71.9 | #144 | 1 |
| [Long Context](https://noometry.com/best/long-context) | 41.6 | #137 | 1 |
| [Writing & Preference](https://noometry.com/best/writing) | 55.7 | #142 | 3 |

## Strengths and weaknesses

Categories where GPT-5.4 nano places highest and lowest among the models ranked in each, with its score against that category's median.

### Strongest categories

GPT-5.4 nano: strongest categories
| Category | Score | vs median | Rank |
| --- | --- | --- | --- |
| [Coding](https://noometry.com/best/coding) | 43.6 | +4.9 | #84 of 340, top 25% |
| [Math](https://noometry.com/best/math) | 40.9 | +4.3 | #88 of 327, top 27% |
| [Knowledge](https://noometry.com/best/knowledge) | 41.9 | +4.6 | #103 of 314, top 33% |

### Weakest categories

GPT-5.4 nano: weakest categories
| Category | Score | vs median | Rank |
| --- | --- | --- | --- |
| [Multimodal](https://noometry.com/best/multimodal) | 36.7 | −1.8 | #78 of 128, top 61% |
| [Reasoning](https://noometry.com/best/reasoning) | 23.7 | +0.1 | #173 of 350, top 50% |
| [Instruction Following](https://noometry.com/best/instruction-following) | 71.9 | +0.7 | #144 of 305, top 48% |

## Closest competitors

The models ranked just above and below GPT-5.4 nano. When scores are this close, price and speed are often the better way to choose.

Models ranked closest to GPT-5.4 nano
| Model | Rank | Score | Blended $/M | Speed |  |
| --- | --- | --- | --- | --- | --- |
| [Solar Pro4](https://noometry.com/models/solar-pro4) | #121 | 42.1 | $0.52 | — | [Compare](https://noometry.com/compare/gpt-5-4-nano-vs-solar-pro4) |
| [GLM-4.5](https://noometry.com/models/glm-4-5) | #122 | 42.0 | $1 | 32 | [Compare](https://noometry.com/compare/glm-4-5-vs-gpt-5-4-nano) |
| [Qwen3.5 35B-A3B](https://noometry.com/models/qwen3-5-35b-a3b) | #123 | 42.0 | $0.69 | — | [Compare](https://noometry.com/compare/gpt-5-4-nano-vs-qwen3-5-35b-a3b) |
| [GLM-4.7](https://noometry.com/models/glm-4-7) | #124 | 42.0 | $1 | — | [Compare](https://noometry.com/compare/glm-4-7-vs-gpt-5-4-nano) |
| [Amazon Nova Experimental Chat 10 09](https://noometry.com/models/amazon-nova-experimental-chat-10-09) | #126 | 41.9 | — | — | [Compare](https://noometry.com/compare/amazon-nova-experimental-chat-10-09-vs-gpt-5-4-nano) |
| [Qwen3.5 27B](https://noometry.com/models/qwen3-5-27b) | #127 | 41.9 | $0.82 | — | [Compare](https://noometry.com/compare/gpt-5-4-nano-vs-qwen3-5-27b) |
| [GPT-5 Mini](https://noometry.com/models/gpt-5-mini) | #128 | 41.8 | $0.69 | 3 | [Compare](https://noometry.com/compare/gpt-5-4-nano-vs-gpt-5-mini) |
| [ERNIE 5.0 0110](https://noometry.com/models/ernie-5-0) | #129 | 41.8 | — | — | [Compare](https://noometry.com/compare/ernie-5-0-vs-gpt-5-4-nano) |

Sponsored placements are available on pages like this one. [Advertise on Noometry](https://noometry.com/advertise)

## Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

### Coding

GPT-5.4 nano Coding benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [SciCode](https://noometry.com/benchmarks/scicode) | 46.9% | #55 of 121, top 46% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [WeirdML](https://noometry.com/benchmarks/weirdml) | 49.2% | #49 of 119, top 42% | high | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [WeirdML](https://noometry.com/benchmarks/weirdml) | 38% |  | none | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LMArena Coding](https://noometry.com/benchmarks/arena-coding) | 1405 | #134 of 294, top 46% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [ALE-Bench](https://noometry.com/benchmarks/ale-bench) | 1,005 | #42 of 105, top 40% | high | [Epoch AI](https://epoch.ai/benchmarks) |  |

### Reasoning

GPT-5.4 nano Reasoning benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [ARC-AGI-2](https://noometry.com/benchmarks/arc-agi-2) | 3.6% |  | high | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [ARC-AGI-2](https://noometry.com/benchmarks/arc-agi-2) | 1.5% |  | low | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [ARC-AGI-2](https://noometry.com/benchmarks/arc-agi-2) | 1.9% |  | medium | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [ARC-AGI-2](https://noometry.com/benchmarks/arc-agi-2) | 5.7% | #54 of 83, top 66% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [Kagi LLM Benchmark](https://noometry.com/benchmarks/kagi-reasoning) | 39.7% | #80 of 99, top 81% |  | [Kagi LLM Benchmark](https://help.kagi.com/kagi/ai/llm-benchmark.html) |  |
| [ARC-AGI-1](https://noometry.com/benchmarks/arc-agi-1) | 38.2% |  | high | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [ARC-AGI-1](https://noometry.com/benchmarks/arc-agi-1) | 18.3% |  | low | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [ARC-AGI-1](https://noometry.com/benchmarks/arc-agi-1) | 33% |  | medium | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [ARC-AGI-1](https://noometry.com/benchmarks/arc-agi-1) | 51.5% | #56 of 83, top 68% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [CritPt](https://noometry.com/benchmarks/critpt) | 9.3% | #49 of 134, top 37% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [Chess Puzzles](https://noometry.com/benchmarks/chess-puzzles) | 30% | #36 of 129, top 28% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-04-14 |
| [Chess Puzzles](https://noometry.com/benchmarks/chess-puzzles) | 17% |  | low | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-07 |
| [Chess Puzzles](https://noometry.com/benchmarks/chess-puzzles) | 3% |  | none | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-07 |
| [LMArena Hard Prompts](https://noometry.com/benchmarks/arena-hard-prompts) | 1381 | #137 of 297, top 47% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [Mystery Game Puzzles](https://noometry.com/benchmarks/mystery-game-puzzles) | 5% |  | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-27 |
| [Mystery Game Puzzles](https://noometry.com/benchmarks/mystery-game-puzzles) | 3% |  | low | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-27 |
| [Mystery Game Puzzles](https://noometry.com/benchmarks/mystery-game-puzzles) | 6% |  | medium | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-27 |
| [Mystery Game Puzzles](https://noometry.com/benchmarks/mystery-game-puzzles) | 9% | #61 of 74, top 83% | none | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-27 |
| [DTBench](https://noometry.com/benchmarks/dtbench) | 80.3% | #73 of 151, top 49% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LMCA](https://noometry.com/benchmarks/lmca) | 36.9% | #60 of 125, top 48% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [Epoch Capabilities Index](https://noometry.com/benchmarks/epoch-capabilities-index) | 145.81 | #79 of 213, top 38% |  | [Epoch AI](https://epoch.ai/eci) | 2026-03-17 |
| [ForecastBench](https://noometry.com/benchmarks/forecastbench) | 57.3 | #59 of 72, top 82% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |

### Math

GPT-5.4 nano Math benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [FrontierMath (Tiers 1-3)](https://noometry.com/benchmarks/frontiermath) | 44.9% | #50 of 81, top 62% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-12 |
| [FrontierMath (Tiers 1-3)](https://noometry.com/benchmarks/frontiermath) | 20.4% |  | low | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-28 |
| [FrontierMath (Tiers 1-3)](https://noometry.com/benchmarks/frontiermath) | 4.6% |  | none | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-28 |
| [FrontierMath Tier 4](https://noometry.com/benchmarks/frontiermath-tier-4) | 12.2% | #50 of 63, top 80% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-12 |
| [OTIS Mock AIME 2024-2025](https://noometry.com/benchmarks/otis-mock-aime) | 87.8% | #60 of 173, top 35% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-04-14 |
| [OTIS Mock AIME 2024-2025](https://noometry.com/benchmarks/otis-mock-aime) | 68.9% |  | low | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-07 |
| [OTIS Mock AIME 2024-2025](https://noometry.com/benchmarks/otis-mock-aime) | 46.7% |  | none | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-07 |
| [ProofBench](https://noometry.com/benchmarks/proofbench) | 5% | #69 of 77, top 90% | high | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LMArena Math](https://noometry.com/benchmarks/arena-math) | 1406 | #120 of 285, top 43% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [FrontierMath (Feb 2025 set)](https://noometry.com/benchmarks/frontiermath-2025-02) | 25.9% | #24 of 68, top 36% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-04-15 |
| [FrontierMath Tier 4 (v1)](https://noometry.com/benchmarks/frontiermath-tier-4-v1) | 6.3% | #23 of 55, top 42% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-04-15 |

### Knowledge

GPT-5.4 nano Knowledge benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [GPQA Diamond](https://noometry.com/benchmarks/gpqa-diamond) | 78.5% | #84 of 186, top 46% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-04-14 |
| [GPQA Diamond](https://noometry.com/benchmarks/gpqa-diamond) | 72.2% |  | low | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-07 |
| [GPQA Diamond](https://noometry.com/benchmarks/gpqa-diamond) | 55.6% |  | none | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-07 |
| [SimpleQA Verified](https://noometry.com/benchmarks/simpleqa-verified) | 11.7% | #73 of 77, top 95% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-27 |
| [Vectara Hallucination Rate](https://noometry.com/benchmarks/vectara-hallucination) (lower is better) | 3.1% | Best of 96 |  | [Vectara Hallucination Leaderboard](https://github.com/vectara/hallucination-leaderboard) |  |
| [LMArena Expert](https://noometry.com/benchmarks/arena-expert) | 1396 | #126 of 273, top 47% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |

### Multimodal

GPT-5.4 nano Multimodal benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [LMArena Vision](https://noometry.com/benchmarks/arena-vision) | 1196 | #80 of 122, top 66% | high | [LMArena](https://lmarena.ai/leaderboard/vision) | 2026-10-09 |

### Multilingual

GPT-5.4 nano Multilingual benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [LMArena Non-English](https://noometry.com/benchmarks/arena-non-english) | 1359 | #140 of 297, top 48% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Chinese](https://noometry.com/benchmarks/arena-chinese) | 1392 | #142 of 285, top 50% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena French](https://noometry.com/benchmarks/arena-french) | 1396 | #123 of 223, top 56% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena German](https://noometry.com/benchmarks/arena-german) | 1367 | #118 of 231, top 52% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Japanese](https://noometry.com/benchmarks/arena-japanese) | 1343 | #102 of 211, top 49% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Korean](https://noometry.com/benchmarks/arena-korean) | 1320 | #119 of 213, top 56% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Russian](https://noometry.com/benchmarks/arena-russian) | 1363 | #139 of 283, top 50% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Spanish](https://noometry.com/benchmarks/arena-spanish) | 1371 | #133 of 226, top 59% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |

### Instruction Following

GPT-5.4 nano Instruction Following benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [LMArena Instruction Following](https://noometry.com/benchmarks/arena-instruction-following) | 1362 | #138 of 298, top 47% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |

### Long Context

GPT-5.4 nano Long Context benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [LMArena Longer Query](https://noometry.com/benchmarks/arena-longer-query) | 1366 | #144 of 291, top 50% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |

### Writing & Preference

GPT-5.4 nano Writing & Preference benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [LMArena Text](https://noometry.com/benchmarks/arena-text) | 1372 | #143 of 297, top 49% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Creative Writing](https://noometry.com/benchmarks/arena-creative-writing) | 1314 | #160 of 295, top 55% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Multi-Turn](https://noometry.com/benchmarks/arena-multi-turn) | 1382 | #133 of 295, top 46% | high | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |

## API pricing by provider

GPT-5.4 nano API prices
| Route | Input $/M | Output $/M | Cached input $/M | Checked |
| --- | --- | --- | --- | --- |
| [azure](https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/models) | $0.20 | $1.25 | $0.02 | 2026-10-10 |
| [openai](https://platform.openai.com/docs/models) | $0.20 | $1.25 | $0.02 | 2026-10-10 |
| [openrouter](https://openrouter.ai/openai/gpt-5.4-nano) | $0.20 | $1.25 | $0.02 | 2026-10-10 |

[All OpenAI API prices →](https://noometry.com/llm-pricing/openai) [Estimate your cost →](https://noometry.com/tools/cost-calculator)

## Compare GPT-5.4 nano

-   [GPT-5.4 nano vs GPT-5 Nano](https://noometry.com/compare/gpt-5-4-nano-vs-gpt-5-nano)
-   [GPT-5.4 nano vs GLM-4.7](https://noometry.com/compare/glm-4-7-vs-gpt-5-4-nano)
-   [GPT-5.4 nano vs Amazon Nova Experimental Chat 10 09](https://noometry.com/compare/amazon-nova-experimental-chat-10-09-vs-gpt-5-4-nano)
-   [GPT-5.4 nano vs Qwen3.5 35B-A3B](https://noometry.com/compare/gpt-5-4-nano-vs-qwen3-5-35b-a3b)
-   [GPT-5.4 nano vs Qwen3.5 27B](https://noometry.com/compare/gpt-5-4-nano-vs-qwen3-5-27b)
-   [GPT-5.4 nano vs GLM-4.5](https://noometry.com/compare/glm-4-5-vs-gpt-5-4-nano)
-   [GPT-5.4 nano vs GPT-5 Mini](https://noometry.com/compare/gpt-5-4-nano-vs-gpt-5-mini)
-   [GPT-5.4 nano vs Claude Fable 5.1](https://noometry.com/compare/claude-fable-5-1-vs-gpt-5-4-nano)
-   [GPT-5.4 nano vs Gemini 3.8 Flash](https://noometry.com/compare/gemini-3-8-flash-vs-gpt-5-4-nano)
-   [GPT-5.4 nano vs Kimi K3](https://noometry.com/compare/gpt-5-4-nano-vs-kimi-k3)
-   [GPT-5.4 nano vs Grok 4.6](https://noometry.com/compare/gpt-5-4-nano-vs-grok-4-6)
-   [GPT-5.4 nano vs Qwen3.8 Max](https://noometry.com/compare/gpt-5-4-nano-vs-qwen3-8-max)
-   [GPT-5.4 nano vs GLM-5.3](https://noometry.com/compare/glm-5-3-vs-gpt-5-4-nano)
-   [GPT-5.4 nano vs Muse Spark 1.3](https://noometry.com/compare/gpt-5-4-nano-vs-muse-spark-1-3)

## Other OpenAI models

-   [GPT-6 Astra](https://noometry.com/models/gpt-6-astra)70.8
-   [GPT-6.1 Sol](https://noometry.com/models/gpt-6-1-sol)65.6
-   [GPT-5.6 Sol](https://noometry.com/models/gpt-5-6-sol)65.0
-   [GPT-5.5 Pro](https://noometry.com/models/gpt-5-5-pro)64.3
-   [GPT-5.5](https://noometry.com/models/gpt-5-5)63.4
-   [GPT-6 Sol](https://noometry.com/models/gpt-6-sol)61.8
-   [GPT-5.4](https://noometry.com/models/gpt-5-4)59.4
-   [GPT-5.6 Terra](https://noometry.com/models/gpt-5-6-terra)59.2

## Frequently asked questions

### How good is GPT-5.4 nano?

GPT-5.4 nano by OpenAI ranks 125th of 354 ranked models on the Noometry Index as of October 2026, with a score of 41.9. Its strongest category is multimodal, where it ranks 78th. API pricing starts at $0.20 per million input tokens and $1.25 per million output tokens, with a 400K-token context window.

### How much does GPT-5.4 nano cost?

GPT-5.4 nano costs $0.20 per million input tokens and $1.25 per million output tokens on OpenAI's own API, with cached input at $0.02.

### What is GPT-5.4 nano's context window?

GPT-5.4 nano accepts up to 400K tokens of input and can write up to 128K tokens in one response.

### Is GPT-5.4 nano open source?

No. GPT-5.4 nano is proprietary and available only through OpenAI's API and partner platforms.

### How fast is GPT-5.4 nano?

GPT-5.4 nano generated about 19 output tokens per second in the Kagi LLM Benchmark's timed runs. Speed varies by provider, load and reasoning effort.

### What are GPT-5.4 nano's strengths and weaknesses?

Relative to other ranked models, GPT-5.4 nano places best in coding, math, knowledge and lowest in multimodal, reasoning, instruction following.

### What is GPT-5.4 nano best at?

Its best category is multimodal, where it ranks 78th on Noometry.

### Cite this page

Noometry. (2026). GPT-5.4 nano benchmarks and pricing. Retrieved October 10, 2026, from https://noometry.com/models/gpt-5-4-nano

Quote Noometry with a link back to this page. It is also available in [Markdown](https://noometry.com/md/models/gpt-5-4-nano.md).