UnifyBench

RankingEffort benchmarks

GPT-5.3 Codex · XHigh

Compare models

12 measured results · 1 effort settings · 12 benchmark/harness combinations

Model evidence and sources →

Change model

Compare with

Choose a second model and its effort to see both profiles and benchmark differences below.

Compare with

Capability profile

One configuration, seven capabilities

Comparison score · 0–100
GPT-5.3 Codex · XHigh capability radarComparison scores from 0 to 100. Unknown capabilities have no point; lines stop at gaps. Hollow points are preliminary. Exact values and support labels follow the chart.AgenticHard reasoningCodingHuman prefKnowledgeMultimodalLong context50100GPT-5.3 Codex · XHigh · Agentic: 19.9 · SupportedGPT-5.3 Codex · XHigh · Hard reasoning: 45.7 · SupportedGPT-5.3 Codex · XHigh · Coding: 25.8 · SupportedGPT-5.3 Codex · XHigh · Knowledge: 41.7 · SupportedGPT-5.3 Codex · XHigh · Multimodal: 34.9 · PreliminaryGPT-5.3 Codex · XHigh · Long context: 68.6 · Preliminary

Gaps are unknown, not zero. Hollow points are preliminary.

  • Agentic

    A19.9Supported

  • Hard reasoning

    A45.7Supported

  • Coding

    A25.8Supported

  • Human pref

    Unknown

  • Knowledge

    A41.7Supported

  • Multimodal

    A34.9Preliminary

  • Long context

    A68.6Preliminary

Each profile belongs to the named configuration. Capability scores use a shared panel for all axes, distinct from the overall ranking panel; they are not accuracy percentages or direct head-to-head wins.

Capability score evidence

GPT-5.3 Codex · XHigh: 18.3 overall comparison score. Meets the overall evidence requirements.

6 areas measured. The available task mix affects this estimate; missing areas remain unknown.

Which sources carry the weight?
  • Artificial Analysis · 81.6% direct comparison weight · independent of the model provider
  • SWE-rebench · 18.4% direct comparison weight · independent of the model provider
Which benchmark families carry the weight?
  • swe-rebench · 18.4% direct comparison weight
  • IFBench · 8.6% direct comparison weight
  • Mmmu Pro · 8.5% direct comparison weight
  • Humanity’s Last Exam · 8.5% direct comparison weight
  • Critpt · 8.4% direct comparison weight
  • Other families (6) · 47.6% direct comparison weight

These shares describe direct comparison weight, not fractions of the final score or confidence. Source balancing does not give every family equal weight or remove correlations between overlapping sources.

How sensitive is this score?

The audit removes one source at a time and refits against the same overall reference panel. These published results use the matching dataset audited 2026-10-20.

Both ranks use the same surviving configurations for each removal. The cohort starts with 444 configurations supported in the original audit, including separate effort settings. These are diagnostic positions, not the default leaderboard ranks. Eligibility is not rechecked after removal.

What changes when a source is removed?
Removed sourceRank before → afterRank movementCohort remaining / lost supportScore after (change)
Artificial Analysis126 → 181Down 55203 remaining / 241 lost1.9 (-16.4 points)
ARC Prize147 → 162Down 15444 remaining / 0 lost18.1 (-0.2 points)
Datacurve147 → 144Up 3444 remaining / 0 lost15.4 (-2.9 points)
LiveCodeBench147 → 147No movement444 remaining / 0 lost18.2 (-0.1 points)
Scale147 → 146Up 1444 remaining / 0 lost20.4 (+2.1 points)
Terminal-Bench147 → 146Up 1444 remaining / 0 lost18.2 (-0.1 points)
Cognition147 → 145Up 2444 remaining / 0 lost18.1 (-0.2 points)
LiveBench146 → 151Down 5443 remaining / 1 lost18.2 (-0.1 points)
SWE-rebench147 → 111Up 36444 remaining / 0 lost30.2 (+11.9 points)
Epoch AI144 → 143Up 1431 remaining / 13 lost18.8 (+0.6 points)
Vals AI147 → 143Up 4443 remaining / 1 lost17.7 (-0.6 points)
Provider reports · Anthropic147 → 147No movement444 remaining / 0 lost18.3 (-0.0 points)
Provider reports · Moonshot AI147 → 147No movement444 remaining / 0 lost17.9 (-0.4 points)
Provider reports · Meta147 → 147No movement444 remaining / 0 lost18.2 (-0.0 points)
Provider reports · Mistral147 → 147No movement444 remaining / 0 lost18.3 (+0.0 points)
Provider reports · OpenAI147 → 147No movement444 remaining / 0 lost18.4 (+0.1 points)
All provider reports147 → 147No movement444 remaining / 0 lost18.0 (-0.3 points)

Lost support means a configuration no longer has a comparison path to the complete reference panel. Its rank is unavailable. A surviving path alone does not establish leaderboard eligibility.

Overlapping sources may remain correlated even with source balancing. Leave-one-source-out movement is a sensitivity diagnostic, not a spread of source-specific ranks or a confidence measure. Scoring methodology · Dated configuration audit

Every benchmark family
Removed evidenceRecomputed scoreChange
swe-rebench30.2+11.9
deepswe15.4-2.9
tau-bench19.8+1.5
omniscience16.9-1.4
briefcase19.5+1.2
aa-lcr17.1-1.2
mmmu-pro19.3+1.1
arc-agi17.6-0.7
simpleqa19.0+0.7
epoch-game-puzzles17.6-0.7
gpqa17.7-0.6
hle18.8+0.5
critpt17.8-0.5
mmlu-pro18.7+0.4
livebench-instructions18.6+0.4
gdp-pdf18.6+0.3
multi-swe-bench18.0-0.3
arc-agi-318.0-0.3
swe-atlas-qna18.0-0.2
apex-agents18.5+0.2
frontiercode18.1-0.2
terminal-bench-science18.5+0.2
livebench-language18.4+0.2
harvey18.1-0.2
livebench-coding18.1-0.2
analyst-agent18.4+0.1
livebench-reasoning18.1-0.1
terminal-bench18.2-0.1
enigma-eval18.2-0.1
vibe-code18.2-0.1
aime18.2-0.1
gmmlu18.3+0.1
itbench18.2-0.1
vals-finance-agent18.2-0.1
gdpval18.3+0.1
vals-legal-research18.2-0.1
vals-code-migration18.2-0.1
vals-tax-agent18.2-0.0
swe-atlas-test-writing18.2-0.0
math50018.3+0.0
automationbench18.2-0.0
livecodebench18.2-0.0
vals-excel-modeling18.2-0.0
enterprise-ops18.3+0.0
livebench-math18.3+0.0
frontiermath18.3+0.0
ifbench18.3+0.0
livebench-data18.3+0.0
swe-atlas-refactoring18.3-0.0
swe-bench-pro18.3+0.0
What evidence is missing from the fit?

Across all collected settings for this model: 12 observations, 11 contributing, 0 matched without graph weight and 1 excluded. Counts do not establish rank eligibility.

  • 1 · Benchmark family or source protocol has not been reviewed
Inspect every observation and exclusion →

Benchmark results

These are published benchmark scores. Supported effort estimates also appear on the Capability leaderboard. Results from different harnesses occupy different rows. Missing cells stay unknown; no scores are copied between effort levels.

Benchmark / sourceXHigh
AA-Briefcase EloArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

AA-Briefcase v1.1 is an agentic knowledge work benchmark developed by Artificial Analysis. AA-Briefcase Elo is a combined metric that aggregates rubric pass rate, analytical quality Elo and presentation Elo · Higher is better · Evaluation results measured independently by Artificial Analysis

Unit: elo. Source collected 2026-10-18.

868.34 Elo
Reported settings

GPT-5.3 Codex (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-3-codex

AA-LCR v1.1Artificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-10-18.

83.33%
Reported settings

GPT-5.3 Codex (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-3-codex

CritPtArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Benchmark developed by Argonne and UIUC, with contributions from 60+ researchers globally. CritPt is under review by its authors following external feedback · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-10-18.

16.86%
Reported settings

GPT-5.3 Codex (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-3-codex

GPQA DiamondArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-10-18.

91.52%
Reported settings

GPT-5.3 Codex (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-3-codex

Humanity's Last ExamArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-10-18.

42.49%
Reported settings

GPT-5.3 Codex (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-3-codex

IFBenchArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-10-18.

75.37%
Reported settings

GPT-5.3 Codex (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-3-codex

MMMU-ProArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-10-18.

78.5%
Reported settings

GPT-5.3 Codex (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-3-codex

AA-Omniscience IndexArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct. · Evaluation results measured independently by Artificial Analysis

Unit: index. Source collected 2026-10-18.

10.87
Reported settings

GPT-5.3 Codex (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-3-codex

AA-Omniscience AccuracyArtificial Analysis · 6000 questions; 1 attempt; correct-answer share; GPT-5.6 Luna medium grader
Benchmark details

AA-Omniscience Accuracy (higher is better) measures the proportion of correctly answered questions out of all questions, regardless of whether the model chooses to answer · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-10-18.

52.88%
Reported settings

GPT-5.3 Codex (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-3-codex

𝜏²-Bench TelecomArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-10-18.

85.96%
Reported settings

GPT-5.3 Codex (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-3-codex

Terminal-Bench HardArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-10-18.

53.03%
Reported settings

GPT-5.3 Codex (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-3-codex

SWE-rebench · 2026-02 · toolsSWE-rebench · SWE-rebench standardized ReAct tools
Benchmark details

Average resolved rate across repeated runs, not best-of-N success. Disjoint complete calendar months within each model’s tested range; standardized harness and all-language task set.

Unit: percent. Source collected 2026-10-18.

58.6%
Reported settings

SWE-rebench standardized ReAct tools; swe-rebench-2026-02-tools; gpt-5.3-codex-xhigh; checkpoint=gpt-5.3-codex

What these comparisons mean

Each source reports its own effort labels. Matching labels do not establish equal compute budgets or identical fallback behavior. Open Reported settings to inspect the exact model name, agent and task coverage. Multiple results at one level remain visible rather than selecting the best. Source-reported None or Non-reasoning labels are not treated as evidence that an API supports those settings.

These measurements use their own dated source collection. The leaderboard fits identified configurations jointly; it contains no pooled model ratings. Explicit thinking budgets are separate configurations. Read the methodology.

Download all effort benchmark results · 8,915 results across 259 models · Collected 2026-10-18