# GPT-5.3 CodexHigh

> Source: Benchize (https://benchize.com) — https://benchize.com/models/gpt-5-3-codex-high. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/gpt-5-3-codex-high.

OpenAI

Canonical URL: https://benchize.com/models/gpt-5-3-codex-high
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 | Latency | Source |  |
| --- | --- | --- | --- | --- | --- | --- | --- |
| [SWE-bench Verified (Epoch AI)](https://benchize.com/benchmarks/epoch-ai%3Aswe_bench_verified.md) | 74.79% | SE ±1.98 pp | 8 / 10 | — | — | [epoch ai · SWE-Bench verified](https://epoch.ai/data/benchmark_data.zip#file=swe_bench_verified.csv) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.7479338842975206 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.7479338842975206 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion SWE-Bench verified reasoningBudget Not reported Reasoning effort high sourceModelVersion gpt-5.3-codex_high ### Reported statistics Uncertainty Standard error: ±1.98 pp standardError 0.019756739501473306 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=swe_bench_verified.csv) Source epoch ai Original model label gpt-5.3-codex Source version SWE-Bench verified Evaluated 2026-02-25T10:37:06.792Z Observed 2026-10-22T03:00:05.281Z Evidence SHA-256 `6480176a5d9b1573676fcacc5df5e8c58f9e5fce34025600a19b3bf60617561c` Source locator `/rows/11` |  |  |  |  |  |  |  |
| [Terminal-Bench 2.0 / easy](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-2%3Aeasy.md) | 100% | SE ±0 pp | 1 / 23 | $7.68 | 11,264.79source duration | [Vals AI · 2.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2.T48tPR4O.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 100 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 7.677245 \| USD \| per_test \| Lower is better \| \| `latency` \| 11264.79 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort high Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 67 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2.T48tPR4O.json) Source Vals AI Original model label openai/gpt-5.3-codex Source version 2.0 Observed 2026-10-22T02:10:29.313Z Evidence SHA-256 `38e640d995259778e531f048b2f2c7b3b281cb05b56be643ddc9dc53eb85adf7` Source locator `/tasks/easy/openai~1gpt-5.3-codex` |  |  |  |  |  |  |  |
| [Terminal-Bench 2.0 / medium](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-2%3Amedium.md) | 69.09% | SE ±6.29 pp | 6 / 23 | $0.56 | 819.26source duration | [Vals AI · 2.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2.T48tPR4O.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 69.091 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.558345 \| USD \| per_test \| Lower is better \| \| `latency` \| 819.257 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort high Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.29 pp sourcePopulation 67 stderr 6.289 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2.T48tPR4O.json) Source Vals AI Original model label openai/gpt-5.3-codex Source version 2.0 Observed 2026-10-22T02:10:29.313Z Evidence SHA-256 `38e640d995259778e531f048b2f2c7b3b281cb05b56be643ddc9dc53eb85adf7` Source locator `/tasks/medium/openai~1gpt-5.3-codex` |  |  |  |  |  |  |  |
| [Terminal-Bench 2.0 / overall](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-2%3Aoverall.md) | 64.05% | SE ±5.12 pp | 6 / 23 | $0.35 | 506.28source duration | [Vals AI · 2.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2.T48tPR4O.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 64.045 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.345045 \| USD \| per_test \| Lower is better \| \| `latency` \| 506.283 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort high Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±5.12 pp sourcePopulation 67 stderr 5.115 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2.T48tPR4O.json) Source Vals AI Original model label openai/gpt-5.3-codex Source version 2.0 Observed 2026-10-22T02:10:29.313Z Evidence SHA-256 `38e640d995259778e531f048b2f2c7b3b281cb05b56be643ddc9dc53eb85adf7` Source locator `/tasks/overall/openai~1gpt-5.3-codex` |  |  |  |  |  |  |  |
| [BullshitBench v2](https://benchize.com/benchmarks/bullshitbench%3Av2.md) | 20% | — | 133 / 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` \| 0.2 \| score_fraction \| all_attempts \| Higher is better \| \| `red_rate` \| 0.46 \| score_fraction \| all_attempts \| Lower is better \| \| `refusal_rate` \| 0.0 \| score_fraction \| all_attempts \| Lower is better \| \| `avg_score` \| 0.69 \| score \| judge_grade_0_to_2 \| Higher is better \| ### Configuration Reasoning effort high suite v2 thinkingEnabled true ### Reported statistics acceptedCount 46 answeredCount 100 clearCount 20 errorCount 0 partialCount 34 rank 155 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-5.3-codex Source version v2 Observed 2026-10-22T03:10:12.603Z Evidence SHA-256 `98b485e08764d1c97144bc4f7b28ced1a96d50850947c775870dfda0e5d749b5` Source locator `/rows/154` |  |  |  |  |  |  |  |
| [Terminal-Bench 2.0 / hard](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-2%3Ahard.md) | 50% | SE ±9.29 pp | 6 / 23 | $1.02 | 1,501.97source duration | [Vals AI · 2.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2.T48tPR4O.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 50 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.023633 \| USD \| per_test \| Lower is better \| \| `latency` \| 1501.972 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort high Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±9.29 pp sourcePopulation 67 stderr 9.285 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2.T48tPR4O.json) Source Vals AI Original model label openai/gpt-5.3-codex Source version 2.0 Observed 2026-10-22T02:10:29.313Z Evidence SHA-256 `38e640d995259778e531f048b2f2c7b3b281cb05b56be643ddc9dc53eb85adf7` Source locator `/tasks/hard/openai~1gpt-5.3-codex` |  |  |  |  |  |  |  |

---
Cite as: GPT-5.3 CodexHigh — Benchize, https://benchize.com/models/gpt-5-3-codex-high (retrieved 2026-10-22)