# GPT-5.4 MiniExtra high

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

OpenAI

Canonical URL: https://benchize.com/models/gpt-5-4-mini-xhigh
Last updated: Thu, 22 Oct 2026 05:40:11 GMT

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

OpenAI

## Benchmarks

| Benchmark | Score | ± | Rank | Cost | Source |  |
| --- | --- | --- | --- | --- | --- | --- |
| [MCP Atlas](https://benchize.com/benchmarks/scale%3Amcp_atlas.md) | 56.7source score | CI 53.6–59.8 | 19 / 22 | — | [scale](https://labs.scale.com/leaderboard/mcp_atlas) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 56.7 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort xhigh Model configuration gpt-5.4-mini (xhigh) sourceModelVersion Not reported thinkingEnabled true ### Reported statistics Uncertainty Confidence interval: 53.6–59.8 contaminationWarning Not reported rank 31 reportedConfidenceHalfWidth 3.1 ### Source [Original source](https://labs.scale.com/leaderboard/mcp_atlas) Source scale Original model label gpt-5.4-mini (xhigh) Observed 2026-10-22T03:30:07.875Z Evidence SHA-256 `7b0bc178e35a1688340e9efca5795a8f54d07e56a98f9cacd5e4471e0da266e9` Source locator `/rows/30` |  |  |  |  |  |  |
| [ARC-AGI-2](https://benchize.com/benchmarks/arc-prize%3Av2_Semi_Private.md) | 18.9% | — | 132 / 243 | 0.75usd per task | [arc prize · v2](https://arcprize.org/media/data/leaderboard/v2.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.189 \| score_fraction \| v2_Semi_Private \| Higher is better \| \| `cost_per_task` \| 0.75 \| usd_per_task \| v2_Semi_Private \| Lower is better \| ### Configuration evaluationSplit v2_Semi_Private harness Not reported modelGroup gpt-5-4-mini-reasoning modelType CoT Reasoning effort xhigh sourceModelVersion gpt-5-4-mini-xhigh thinkingEnabled true ### Source [Original source](https://arcprize.org/media/data/leaderboard/v2.json) Source arc prize Original model label GPT-5.4 Mini (XHigh) Source version v2 Observed 2026-10-22T05:20:04.306Z Evidence SHA-256 `4351eb696e78113b0459e9bb5f0258318b905efef34f473dcea51ce641b0d94d` Source locator `/evaluations/135` |  |  |  |  |  |  |
| [FrontierCode 1 main](https://benchize.com/benchmarks/frontiercode%3Av1%3Amain.md) | 17.82% | — | 22 / 39 | $1.67 | [frontiercode](https://cognition.com/data/frontiercode-leaderboard/data.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.1782 \| score_fraction \| blocking_criteria_adjusted \| Higher is better \| \| `correct` \| 0.2 \| score_fraction \| raw_rubric \| Higher is better \| \| `cost` \| 1.6706 \| USD \| per_rollout \| Lower is better \| ### Configuration harness codex Reasoning effort xhigh sourceVersion v1 subset main ### Reported statistics flaggedRate Not reported sampleSize 100 ### Source [Original source](https://cognition.com/data/frontiercode-leaderboard/data.json) Source frontiercode Original model label GPT-5.4-mini Observed 2026-10-22T05:40:03.220Z Evidence SHA-256 `ff0040778b7bdf69c705ad7474f771fb5e53cfaf5cf8573b3a4614153543d75f` Source locator `/v1/data/GPT-5.4-mini/xhigh/main` |  |  |  |  |  |  |
| [FrontierCode 1.1 extended](https://benchize.com/benchmarks/frontiercode%3Av1_1%3Aextended.md) | 43.01% | — | 106 / 125 | $1.26 | [frontiercode](https://cognition.com/data/frontiercode-leaderboard/data.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.4301 \| score_fraction \| blocking_criteria_adjusted \| Higher is better \| \| `correct` \| 0.4794 \| score_fraction \| raw_rubric \| Higher is better \| \| `cost` \| 1.2621 \| USD \| per_rollout \| Lower is better \| ### Configuration harness codex Reasoning effort xhigh sourceVersion v1_1 subset extended ### Reported statistics flaggedRate 0 sampleSize 150 ### Source [Original source](https://cognition.com/data/frontiercode-leaderboard/data.json) Source frontiercode Original model label GPT-5.4-mini Observed 2026-10-22T05:40:03.220Z Evidence SHA-256 `ff0040778b7bdf69c705ad7474f771fb5e53cfaf5cf8573b3a4614153543d75f` Source locator `/v1_1/data/GPT-5.4-mini/xhigh/extended` |  |  |  |  |  |  |
| [ARC-AGI-1](https://benchize.com/benchmarks/arc-prize%3Av1_Semi_Private.md) | 63.67% | — | 142 / 245 | 0.51usd per task | [arc prize · v1](https://arcprize.org/media/data/leaderboard/v1.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.6367 \| score_fraction \| v1_Semi_Private \| Higher is better \| \| `cost_per_task` \| 0.51 \| usd_per_task \| v1_Semi_Private \| Lower is better \| ### Configuration evaluationSplit v1_Semi_Private harness Not reported modelGroup gpt-5-4-mini-reasoning modelType CoT Reasoning effort xhigh sourceModelVersion gpt-5-4-mini-xhigh thinkingEnabled true ### Source [Original source](https://arcprize.org/media/data/leaderboard/v1.json) Source arc prize Original model label GPT-5.4 Mini (XHigh) Source version v1 Observed 2026-10-22T05:20:04.066Z Evidence SHA-256 `2bf68d22dafba37047d07ad833ebc0e8199c9aa2f9eeaed50c75c1b704462836` Source locator `/evaluations/134` |  |  |  |  |  |  |
| [FrontierCode 1 extended](https://benchize.com/benchmarks/frontiercode%3Av1%3Aextended.md) | 36% | — | 23 / 39 | $1.39 | [frontiercode](https://cognition.com/data/frontiercode-leaderboard/data.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.36 \| score_fraction \| blocking_criteria_adjusted \| Higher is better \| \| `correct` \| 0.3987 \| score_fraction \| raw_rubric \| Higher is better \| \| `cost` \| 1.3899 \| USD \| per_rollout \| Lower is better \| ### Configuration harness codex Reasoning effort xhigh sourceVersion v1 subset extended ### Reported statistics flaggedRate Not reported sampleSize 150 ### Source [Original source](https://cognition.com/data/frontiercode-leaderboard/data.json) Source frontiercode Original model label GPT-5.4-mini Observed 2026-10-22T05:40:03.220Z Evidence SHA-256 `ff0040778b7bdf69c705ad7474f771fb5e53cfaf5cf8573b3a4614153543d75f` Source locator `/v1/data/GPT-5.4-mini/xhigh/extended` |  |  |  |  |  |  |
| [FrontierCode 1.1 main](https://benchize.com/benchmarks/frontiercode%3Av1_1%3Amain.md) | 27.04% | — | 105 / 125 | $1.52 | [frontiercode](https://cognition.com/data/frontiercode-leaderboard/data.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.2704 \| score_fraction \| blocking_criteria_adjusted \| Higher is better \| \| `correct` \| 0.308 \| score_fraction \| raw_rubric \| Higher is better \| \| `cost` \| 1.5201 \| USD \| per_rollout \| Lower is better \| ### Configuration harness codex Reasoning effort xhigh sourceVersion v1_1 subset main ### Reported statistics flaggedRate 0 sampleSize 100 ### Source [Original source](https://cognition.com/data/frontiercode-leaderboard/data.json) Source frontiercode Original model label GPT-5.4-mini Observed 2026-10-22T05:40:03.220Z Evidence SHA-256 `ff0040778b7bdf69c705ad7474f771fb5e53cfaf5cf8573b3a4614153543d75f` Source locator `/v1_1/data/GPT-5.4-mini/xhigh/main` |  |  |  |  |  |  |
| [BullshitBench v2](https://benchize.com/benchmarks/bullshitbench%3Av2.md) | 31% | — | 106 / 191 | — | [bullshitbench · v2](https://api.github.com/repos/petergpt/bullshit-benchmark/contents/data/v2/latest/leaderboard.csv?ref=3c655234d15967738f95da696e3f938d4862dcf7) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `green_rate` \| 0.31 \| score_fraction \| all_attempts \| Higher is better \| \| `red_rate` \| 0.22 \| score_fraction \| all_attempts \| Lower is better \| \| `refusal_rate` \| 0.0 \| score_fraction \| all_attempts \| Lower is better \| \| `avg_score` \| 1.0467 \| score \| judge_grade_0_to_2 \| Higher is better \| ### Configuration Reasoning effort xhigh suite v2 thinkingEnabled true ### Reported statistics acceptedCount 22 answeredCount 100 clearCount 31 errorCount 0 partialCount 47 rank 113 refusalCount 0 sampleSize 100 ### Source [Original source](https://api.github.com/repos/petergpt/bullshit-benchmark/contents/data/v2/latest/leaderboard.csv?ref=3c655234d15967738f95da696e3f938d4862dcf7) Source bullshitbench Original model label openai/gpt-5.4-mini Source version v2 Observed 2026-10-22T03:10:12.603Z Evidence SHA-256 `98b485e08764d1c97144bc4f7b28ced1a96d50850947c775870dfda0e5d749b5` Source locator `/rows/112` |  |  |  |  |  |  |

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