UnifyBench

RankingEffort benchmarks

GPT-5.1 · High effort benchmarks

Compare measured Low, Medium, High and other settings. Every number keeps its source.

GPT-5.1

36 measured results · 6 reported effort labels · 20 benchmark/harness combinations. Model evidence and sources →

Capability score evidence

GPT-5.1 · High: 21.7 overall comparison score. Meets the overall evidence requirements.

Which sources carry the weight?
  • Artificial Analysis · 59.3% direct comparison weight · independent
  • ARC Prize · 40.7% direct comparison weight · independent

These shares describe the comparisons entering the fit, not fractions of the final score.

How sensitive is this score?

Remove one evaluator or family and refit the graph against the same overall panel. Values use the matching dataset audited 2026-09-12. These are sensitivity checks, not confidence intervals; ranking eligibility may also change.

Removed evidenceRecomputed scoreChange
Artificial Analysis20.3-1.4
ARC Prize26.2+4.5
Datacurve18.8-2.9
LiveCodeBench21.8+0.1
Scale23.3+1.5
Terminal-Bench21.7-0.0
Cognition22.3+0.6
LiveBench22.0+0.2
SWE-rebench21.9+0.2
Epoch AI21.3-0.5
Vals AI21.0-0.7
Provider reports · Anthropic21.7+0.0
Provider reports · Moonshot AI21.7-0.0
Provider reports · Meta21.7-0.0
Provider reports · Mistral21.7+0.0
All provider reports21.7-0.1
Every benchmark family
Removed evidenceRecomputed scoreChange
arc-agi26.1+4.4
deepswe18.9-2.9
epoch-game-puzzles20.7-1.0
mmmu-pro22.7+1.0
hle22.5+0.8
frontiercode22.3+0.6
simpleqa22.3+0.6
gdpval22.3+0.6
omniscience21.2-0.5
aa-lcr21.2-0.5
arc-agi-321.3-0.5
mmlu-pro22.2+0.5
ifbench21.4-0.3
tau-bench22.1+0.3
multi-swe-bench21.5-0.2
livebench-instructions22.0+0.2
gpqa21.5-0.2
scicode21.9+0.2
gmmlu21.5-0.2
frontiermath21.9+0.2
harvey21.5-0.2
swe-rebench21.9+0.2
livebench-language21.9+0.2
swe-atlas-qna21.6-0.2
livecodebench21.9+0.2
apex-agents21.9+0.1
gdp-pdf21.9+0.1
terminal-bench21.8+0.1
livebench-coding21.6-0.1
vibe-code21.6-0.1
livebench-reasoning21.6-0.1
critpt21.7-0.1
vals-finance-agent21.7-0.1
swe-atlas-test-writing21.7-0.1
aime21.8+0.1
vals-code-migration21.7-0.1
vals-legal-research21.7-0.1
automationbench21.7-0.0
analyst-agent21.8+0.0
swe-atlas-refactoring21.7-0.0
briefcase21.7-0.0
math50021.8+0.0
vals-excel-modeling21.7-0.0
livebench-math21.8+0.0
itbench21.7-0.0
livebench-data21.7+0.0
swe-bench-pro21.7+0.0
enterprise-ops21.7+0.0
enigma-eval21.7+0.0
What evidence is missing from the fit?

Across all collected settings for this model: 54 observations, 34 contributing, 0 matched without graph weight and 20 excluded. Counts do not establish rank eligibility.

  • 2 · Benchmark family or source protocol has not been reviewed
  • 12 · No reviewed effort for this catalog observation; any separately collected effort measurement is counted separately
  • 6 · Supporting evidence outside the reviewed capability core
Inspect every observation and exclusion →

Capability profile

One configuration, seven capabilities

Comparison score · 0–100
GPT-5.1 · High 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-5.1 · High · Agentic: 18.5 · SupportedGPT-5.1 · High · Hard reasoning: 24.8 · SupportedGPT-5.1 · High · Coding: 22.9 · SupportedGPT-5.1 · High · Knowledge: 32.4 · SupportedGPT-5.1 · High · Multimodal: 27.6 · PreliminaryGPT-5.1 · High · Long context: 54.6 · Preliminary

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

  • Agentic

    A18.5Supported

  • Hard reasoning

    A24.8Supported

  • Coding

    A22.9Supported

  • Human pref

    Unknown

  • Knowledge

    A32.4Supported

  • Multimodal

    A27.6Preliminary

  • Long context

    A54.6Preliminary

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.

Compare this configuration →Explore the model evidence profile →

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 / sourceLowMediumHighThinkingNoneNon-reasoning
AIME 2025Artificial 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 2026-09-12.

Not reportedNot reported
94%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reported
38%
Reported settings

GPT-5.1 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-1-non-reasoning

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 2026-09-12.

Not reportedNot reported
80%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reported
45%
Reported settings

GPT-5.1 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-1-non-reasoning

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

Unit: percent. Source collected 2026-09-12.

Not reportedNot reported
4.86%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reported
0%
Reported settings

GPT-5.1 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-1-non-reasoning

GDPval-AA v2 LeaderboardArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Elo rating for performance on real-world work tasks · Anchored to a human baseline of 1,000 · Higher is better · Evaluation results measured independently by Artificial Analysis

Unit: elo. Source collected 2026-09-12.

Not reportedNot reported
929.53 Elo
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reportedNot reported
Global-MMLU-LiteArtificial 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 2026-09-12.

Not reportedNot reported
90.56%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reportedNot reported
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 2026-09-12.

Not reportedNot reported
87.27%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reported
64.34%
Reported settings

GPT-5.1 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-1-non-reasoning

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 2026-09-12.

Not reportedNot reported
28.5%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reported
5.28%
Reported settings

GPT-5.1 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-1-non-reasoning

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 2026-09-12.

Not reportedNot reported
72.86%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reported
43.2%
Reported settings

GPT-5.1 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-1-non-reasoning

LiveCodeBenchArtificial 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 2026-09-12.

Not reportedNot reported
86.77%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reported
49.42%
Reported settings

GPT-5.1 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-1-non-reasoning

MMLU-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 2026-09-12.

Not reportedNot reported
87.04%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reported
80.13%
Reported settings

GPT-5.1 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-1-non-reasoning

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 2026-09-12.

Not reportedNot reported
75.49%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reported
62.37%
Reported settings

GPT-5.1 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-1-non-reasoning

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 2026-09-12.

Not reportedNot reported
5.4
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reported
-34.45
Reported settings

GPT-5.1 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-1-non-reasoning

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 2026-09-12.

Not reportedNot reported
37.72%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reported
29.47%
Reported settings

GPT-5.1 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-1-non-reasoning

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 2026-09-12.

Not reportedNot reported
81.87%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reported
46.49%
Reported settings

GPT-5.1 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-1-non-reasoning

𝜏³-BankingArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Benchmark developed by Sierra Research · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

Not reportedNot reported
15.88%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reportedNot reported
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 2026-09-12.

Not reportedNot reported
45.45%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reported
22.73%
Reported settings

GPT-5.1 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-1-non-reasoning

Terminal-Bench v2.1Artificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Benchmark developed by the Laude Institute and open-source contributors · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

Not reportedNot reported
52.43%
Reported settings

GPT-5.1 (high); Model source: https://artificialanalysis.ai/models/gpt-5-1

Not reportedNot reportedNot reported
ARC-AGI-2ARC Prize · v2 Semi-Private
Benchmark details

Original public board. Configuration and task coverage may differ; inspect source.

Unit: percent. Source collected 2026-09-12.

1.94%
Reported settings

GPT-5.1 (Thinking, Low); Source model ID: gpt-5-1-2025-11-13-thinking-low; 120 tasks

6.53%
Reported settings

GPT-5.1 (Thinking, Medium); Source model ID: gpt-5-1-2025-11-13-thinking-medium; 120 tasks

17.64%
Reported settings

GPT-5.1 (Thinking, High); Source model ID: gpt-5-1-2025-11-13-thinking-high; 120 tasks

Not reported
0.42%
Reported settings

GPT-5.1 (Thinking, None); Source model ID: gpt-5-1-2025-11-13-thinking-none; 120 tasks

Not reported
HLE · text only · Scale official evaluationScale · Scale official evaluation
Benchmark details

Scale text-only public HLE subset; shares the full HLE family budget; automatic answer judging.

Unit: percent. Source collected 2026-09-12.

Not reportedNot reportedNot reported
24.65%
Reported settings

Scale text-only public HLE subset; shares the full HLE family budget; automatic answer judging.; published row 2025-11-26T20:28:20.000Z; reported margin ±1.82 percentage points; Potential contamination warning: This model was evaluated after the public release of HLE, allowing model builder access to the prompts and solutions.

Not reportedNot reported
HLE · full · Scale official evaluationScale · Scale official evaluation
Benchmark details

Scale full public HLE; 2,500 questions; automatic answer judging; temperature 0 when configurable. Public-set contamination caveats retained per result.

Unit: percent. Source collected 2026-09-12.

Not reportedNot reportedNot reported
23.68%
Reported settings

Scale full public HLE; 2,500 questions; automatic answer judging; temperature 0 when configurable. Public-set contamination caveats retained per result.; published row 2025-11-26T20:24:19.000Z; reported margin ±1.67 percentage points; Potential contamination warning: This model was evaluated after the public release of HLE, allowing model builder access to the prompts and solutions.

Not reportedNot reported
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 · 7,424 results across 240 models · Collected 2026-09-12