# GPT-5.4 NanoLow

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

OpenAI

Canonical URL: https://benchize.com/models/gpt-5-4-nano-low
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 | Cost | Source |  |
| --- | --- | --- | --- | --- | --- |
| [ARC-AGI-1](https://benchize.com/benchmarks/arc-prize%3Av1_Semi_Private.md) | 18.33% | 184 / 213 | 0.01usd per task | [arc prize · v1](https://arcprize.org/media/data/leaderboard/v1.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.1833 \| score_fraction \| v1_Semi_Private \| Higher is better \| \| `cost_per_task` \| 0.01 \| usd_per_task \| v1_Semi_Private \| Lower is better \| ### Configuration evaluationSplit v1_Semi_Private harness Not reported modelGroup gpt-5-4-nano-reasoning modelType CoT Reasoning effort low sourceModelVersion gpt-5-4-nano-low thinkingEnabled true ### Source [Original source](https://arcprize.org/media/data/leaderboard/v1.json) Source arc prize Original model label GPT-5.4 Nano (Low) Source version v1 Observed 2026-10-21T16:34:20.378Z Evidence SHA-256 `302d8267c0522bf450ec1deeac4d060e2cf2fb444ba0a4269014d9535de0bd1f` Source locator `/evaluations/141` |  |  |  |  |  |
| [NC Bench](https://benchize.com/benchmarks/nc-bench%3Aoverall.md) | 77.46% | 132 / 144 | — | [nc bench](https://www.nc-bench.com/models) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 77.46 \| percent \| reported \| Higher is better \| \| `tooling` \| 91.46 \| percent \| category \| Higher is better \| \| `creative_writing` \| 80.93 \| percent \| category \| Higher is better \| \| `language` \| 81.87 \| percent \| category \| Higher is better \| \| `utility` \| 91.42 \| percent \| category \| Higher is better \| \| `reasoning` \| 64.43 \| percent \| category \| Higher is better \| \| `text_editing` \| 82.23 \| percent \| category \| Higher is better \| \| `rule_following` \| 31.65 \| percent \| category \| Higher is better \| \| `hallucination` \| 95.71 \| percent \| category \| Higher is better \| ### Configuration Reasoning effort low Model configuration GPT-5.4 Nano (Reasoning, Low) thinkingEnabled true ### Source [Original source](https://www.nc-bench.com/models) Source nc bench Original model label GPT-5.4 Nano (Reasoning, Low) Observed 2026-10-21T16:31:48.370Z Evidence SHA-256 `ec261022458543624b08a136d01f5a2f0666cd52ffb49c64d6ca983e8543d1b5` Source locator `/rows/176` |  |  |  |  |  |
| [ARC-AGI-2](https://benchize.com/benchmarks/arc-prize%3Av2_Semi_Private.md) | 1.53% | 176 / 212 | 0.01usd per task | [arc prize · v2](https://arcprize.org/media/data/leaderboard/v2.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.0153 \| score_fraction \| v2_Semi_Private \| Higher is better \| \| `cost_per_task` \| 0.01 \| usd_per_task \| v2_Semi_Private \| Lower is better \| ### Configuration evaluationSplit v2_Semi_Private harness Not reported modelGroup gpt-5-4-nano-reasoning modelType CoT Reasoning effort low sourceModelVersion gpt-5-4-nano-low thinkingEnabled true ### Source [Original source](https://arcprize.org/media/data/leaderboard/v2.json) Source arc prize Original model label GPT-5.4 Nano (Low) Source version v2 Observed 2026-10-21T16:34:20.750Z Evidence SHA-256 `bca68e12cc9a01954ae21b641844c1ecbabe146cb24b87f0f8e9b1db27433c29` Source locator `/evaluations/143` |  |  |  |  |  |

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