# GPT-5 Pro > Source: Benchize (https://benchize.com) — https://benchize.com/models/gpt-5-pro. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/gpt-5-pro. OpenAI Canonical URL: https://benchize.com/models/gpt-5-pro Last updated: Thu, 22 Oct 2026 07:32:33 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-2](https://benchize.com/benchmarks/arc-prize%3Av2_Semi_Private.md) | 18.33% | — | 134 / 243 | 7.14usd per task | [arc prize · v2](https://arcprize.org/media/data/leaderboard/v2.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.1833 \| score_fraction \| v2_Semi_Private \| Higher is better \| \| `cost_per_task` \| 7.1432 \| usd_per_task \| v2_Semi_Private \| Lower is better \| ### Configuration evaluationSplit v2_Semi_Private harness Not reported modelGroup Not reported modelType CoT sourceModelVersion gpt-5-pro-2025-10-06 ### Source [Original source](https://arcprize.org/media/data/leaderboard/v2.json) Source arc prize Original model label GPT-5 Pro Source version v2 Observed 2026-10-22T05:20:04.306Z Evidence SHA-256 `4351eb696e78113b0459e9bb5f0258318b905efef34f473dcea51ce641b0d94d` Source locator `/evaluations/81` | | | | | | | | [ARC-AGI-1](https://benchize.com/benchmarks/arc-prize%3Av1_Semi_Private.md) | 70.17% | — | 132 / 245 | 4.78usd per task | [arc prize · v1](https://arcprize.org/media/data/leaderboard/v1.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.7017 \| score_fraction \| v1_Semi_Private \| Higher is better \| \| `cost_per_task` \| 4.7785 \| usd_per_task \| v1_Semi_Private \| Lower is better \| ### Configuration evaluationSplit v1_Semi_Private harness Not reported modelGroup Not reported modelType CoT sourceModelVersion gpt-5-pro-2025-10-06 ### Source [Original source](https://arcprize.org/media/data/leaderboard/v1.json) Source arc prize Original model label GPT-5 Pro Source version v1 Observed 2026-10-22T05:20:04.066Z Evidence SHA-256 `2bf68d22dafba37047d07ad833ebc0e8199c9aa2f9eeaed50c75c1b704462836` Source locator `/evaluations/79` | | | | | | | | [Putnam 2025](https://benchize.com/benchmarks/matharena%3Aputnam--putnam_2025.md) | 77.5% | 95% CI 53.87–101.13% | 1 / 2 | 3.83usd per problem | [matharena · putnam--putnam_2025](https://matharena.ai/competition_tables/putnam--putnam_2025) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 77.50 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 3.83 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition putnam--putnam_2025 predicted false sourceModelName GPT-5-Pro sourceModelVersion openai/gpt-5-pro ### Reported statistics Uncertainty 95% confidence interval: 53.87–101.13% confidenceInterval95 23.63 problems 12 ### Source [Original source](https://matharena.ai/competition_tables/putnam--putnam_2025) Source matharena Original model label GPT-5-Pro Source version putnam--putnam_2025 Observed 2026-10-22T05:30:15.320Z Evidence SHA-256 `09151696c0dd9678a24a575c6b5c9198910abdf8578950b1136fe34312e49c6f` Source locator `/table/rows/2` | | | | | | | | [SimpleBench](https://benchize.com/benchmarks/simplebench%3Aavg-at-5.md) | 61.6% | — | 23 / 62 | — | [simplebench](https://simple-bench.com/static/js/leaderboard-data.js) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 61.6 \| percent \| average_at_5 \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration GPT-5 Pro thinkingEnabled Not reported ### Reported statistics rank 32nd ### Source [Original source](https://simple-bench.com/static/js/leaderboard-data.js) Source simplebench Original model label GPT-5 Pro Observed 2026-10-22T03:40:04.693Z Evidence SHA-256 `b2d0285ad3e4e15750ed20d39897e7558d1c05e8962e90c861b06f93c8019dc9` Source locator `/rows/33` | | | | | | | | [Miklós Schweitzer 2025](https://benchize.com/benchmarks/matharena%3Amiklos--2025.md) | 90% | 95% CI 73.69–106.31% | 1 / 1 | 4.09usd per problem | [matharena · miklos--2025](https://matharena.ai/competition_tables/miklos--2025) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 90.00 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 4.09 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition miklos--2025 predicted false sourceModelName GPT-5-Pro sourceModelVersion openai/gpt-5-pro ### Reported statistics Uncertainty 95% confidence interval: 73.69–106.31% confidenceInterval95 16.31 problems 10 ### Source [Original source](https://matharena.ai/competition_tables/miklos--2025) Source matharena Original model label GPT-5-Pro Source version miklos--2025 Observed 2026-10-22T05:30:15.062Z Evidence SHA-256 `748ecb15adaa05bbfed8e75816ead2065a6d06da0a307cca182de0d00568b72b` Source locator `/table/rows/0` | | | | | | | | [PRBench Legal](https://benchize.com/benchmarks/scale%3Aprbench-legal.md) | 49.89source score | CI 49.53–50.25 | 7 / 26 | — | [scale](https://labs.scale.com/leaderboard/prbench-legal) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 49.89389865 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gpt-5-pro sourceModelVersion Not reported thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 49.53–50.25 contaminationWarning Not reported rank 10 reportedConfidenceHalfWidth 0.3589786494 ### Source [Original source](https://labs.scale.com/leaderboard/prbench-legal) Source scale Original model label gpt-5-pro Observed 2026-10-22T03:30:10.599Z Evidence SHA-256 `c086b601b5cff78b7934cfbcf79f15db48cd423c4e62e56a185b447b03c28f3c` Source locator `/rows/9` | | | | | | | | [PRBench Finance](https://benchize.com/benchmarks/scale%3Aprbench-finance.md) | 51.06source score | CI 50.48–51.65 | 6 / 22 | — | [scale](https://labs.scale.com/leaderboard/prbench-finance) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 51.06359778 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gpt-5-pro sourceModelVersion Not reported thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 50.48–51.65 contaminationWarning Not reported rank 8 reportedConfidenceHalfWidth 0.5876640955 ### Source [Original source](https://labs.scale.com/leaderboard/prbench-finance) Source scale Original model label gpt-5-pro Observed 2026-10-22T03:30:10.417Z Evidence SHA-256 `52572a2526d41cba3e52146e6063412a3850897b7608a0745f62baae712ab29b` Source locator `/rows/7` | | | | | | | --- Cite as: GPT-5 Pro — Benchize, https://benchize.com/models/gpt-5-pro (retrieved 2026-10-22)