# GPT-5.4 Mini

> Source: Benchize (https://benchize.com) — https://benchize.com/models/gpt-5-4-mini. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/gpt-5-4-mini.

OpenAI

Canonical URL: https://benchize.com/models/gpt-5-4-mini
Last updated: Thu, 22 Oct 2026 04:30:12 GMT

Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations.

OpenAI

## Benchmarks

| Benchmark | Score | ± | Rank | Source |  |
| --- | --- | --- | --- | --- | --- |
| [Creative Writing v3](https://benchize.com/benchmarks/eq-bench%3Acreative-writing-v3.md) | 1,665elo | — | 36 / 90 | [eq bench · creative-writing-v3](https://eqbench.com/creative_writing.js) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo_score` \| 1665.0 \| elo \| overall \| Higher is better \| \| `creative_writing_score` \| 16.51 \| rubric_score \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v3 sourceModelVersion gpt-5.4-mini ### Source [Original source](https://eqbench.com/creative_writing.js) Source eq bench Original model label gpt-5.4-mini Source version creative-writing-v3 Observed 2026-10-21T16:35:01.421Z Evidence SHA-256 `722ddb132e8d3a5e5c39b8d445f8475f3f7e24893a007260c6f3e3cce5c1ee18` Source locator `/leaderboardDataCreativeWritingV3/40` |  |  |  |  |  |
| [Longform Creative Writing](https://benchize.com/benchmarks/eq-bench%3Alongform-writing.md) | 73% | — | 29 / 93 | [eq bench · longform-writing](https://eqbench.com/creative_writing_longform.js) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `overall_score_100` \| 73.0 \| percent \| overall \| Higher is better \| \| `chapter1_avg` \| 15.81 \| rubric_score \| overall \| Higher is better \| \| `chapter2_avg` \| 15.61 \| rubric_score \| overall \| Higher is better \| \| `chapter3_avg` \| 15.12 \| rubric_score \| overall \| Higher is better \| \| `chapter4_avg` \| 15.25 \| rubric_score \| overall \| Higher is better \| \| `chapter5_avg` \| 15.21 \| rubric_score \| overall \| Higher is better \| \| `chapter6_avg` \| 15.22 \| rubric_score \| overall \| Higher is better \| \| `chapter7_avg` \| 15.05 \| rubric_score \| overall \| Higher is better \| \| `chapter8_avg` \| 15.24 \| rubric_score \| overall \| Higher is better \| \| `final_judgement_avg` \| 13.71 \| rubric_score \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol longform-writing sourceModelVersion gpt-5.4-mini ### Source [Original source](https://eqbench.com/creative_writing_longform.js) Source eq bench Original model label gpt-5.4-mini Source version longform-writing Observed 2026-10-21T16:35:02.575Z Evidence SHA-256 `3ba283ce357ac7620e16db640a961879c30fe3ddc50a27de9a8ed4290784825c` Source locator `/leaderboardDataLongformV3/100` |  |  |  |  |  |
| [Judgemark v4](https://benchize.com/benchmarks/eq-bench%3Ajudgemark-v4.md) | 27.32% | CI 24.87–35.77% | 37 / 38 | [eq bench · judgemark-v4](https://eqbench.com/judgemark-v4.js) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.273221 \| score_fraction \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol judgemark-v4 sourceModelVersion gpt-5.4-mini ### Reported statistics Uncertainty Confidence interval: 24.87–35.77% confidenceIntervalHigh 0.357704 confidenceIntervalLow 0.248656 ### Source [Original source](https://eqbench.com/judgemark-v4.js) Source eq bench Original model label gpt-5.4-mini Source version judgemark-v4 Observed 2026-10-21T16:35:03.901Z Evidence SHA-256 `2b0d42aef4f614dd199dc38399be7b88075a919db52020c3759f190defc92a4d` Source locator `/leaderboardDataJudgemarkV4/39` |  |  |  |  |  |
| [SVGBench Arena](https://benchize.com/benchmarks/svgbench%3Aarena.md) | 1,554 | — | 48 / 191 | [svgbench](https://svgbench.ai/api/leaderboard?shape=2) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1554 \| rating \| reported \| Higher is better \| ### Configuration Reasoning effort unknown sourceModelId model_openai_gpt_5_4_mini ### Reported statistics coverageCount 27 draws 6 judged 102 losses 42 rank 59 ratingDeviation 28 wins 54 ### Source [Original source](https://svgbench.ai/api/leaderboard?shape=2) Source svgbench Original model label GPT-5.4 Mini Observed 2026-10-22T04:30:04.538Z Evidence SHA-256 `5c0a9c869593820d57d2c05f6d416e1b44a740b4835e66c66e9ad8b8edc59e2c` Source locator `/rows/57` |  |  |  |  |  |
| [Judgemark v2.1](https://benchize.com/benchmarks/eq-bench%3Ajudgemark-v2.md) | 81.14% | — | 13 / 39 | [eq bench · judgemark-v2](https://eqbench.com/judgemark-v2.js) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `judgemark_score` \| 81.14 \| percent \| overall \| Higher is better \| \| `stability` \| 0.714 \| score_fraction \| overall \| Higher is better \| \| `separability` \| 0.834 \| score_fraction \| overall \| Higher is better \| \| `human_corr` \| 0.818 \| correlation \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol judgemark-v2 sourceModelVersion gpt-5.4-mini ### Source [Original source](https://eqbench.com/judgemark-v2.js) Source eq bench Original model label gpt-5.4-mini Source version judgemark-v2 Observed 2026-10-21T16:35:03.564Z Evidence SHA-256 `070cc3fbd6f0333dde2b52bec4329ea3fba7eccbf016bbbbef95e0f8b2197781` Source locator `/leaderboardDataV2/46` |  |  |  |  |  |
| [NC Bench](https://benchize.com/benchmarks/nc-bench%3Aoverall.md) | 80.45% | — | 125 / 144 | [nc bench](https://www.nc-bench.com/models) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 80.45 \| percent \| reported \| Higher is better \| \| `tooling` \| 98.55 \| percent \| category \| Higher is better \| \| `creative_writing` \| 88.1 \| percent \| category \| Higher is better \| \| `language` \| 88.75 \| percent \| category \| Higher is better \| \| `utility` \| 79.37 \| percent \| category \| Higher is better \| \| `reasoning` \| 70.07 \| percent \| category \| Higher is better \| \| `text_editing` \| 90.6 \| percent \| category \| Higher is better \| \| `rule_following` \| 46.32 \| percent \| category \| Higher is better \| \| `hallucination` \| 81.85 \| percent \| category \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration GPT-5.4 Mini thinkingEnabled Not reported ### Source [Original source](https://www.nc-bench.com/models) Source nc bench Original model label GPT-5.4 Mini Observed 2026-10-21T16:31:48.370Z Evidence SHA-256 `ec261022458543624b08a136d01f5a2f0666cd52ffb49c64d6ca983e8543d1b5` Source locator `/rows/164` |  |  |  |  |  |

---
Cite as: GPT-5.4 Mini — Benchize, https://benchize.com/models/gpt-5-4-mini (retrieved 2026-10-22)