RankingEffort benchmarks

o3 · Reasoning

Compare models

18 measured results · 4 effort settings · 16 benchmark/harness combinations

Model evidence and sources →

Change model

Model

Current configurationSources

Compare with

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

Compare withModel and configuration

Capability profile

One configuration, seven capabilities

Comparison score · 0–100

Ao3 · Reasoning3 of 7 capabilities supported

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

Agentic
A13.9Preliminary
Hard reasoning
A18.6Supported
Coding
A21.2Supported
Human pref
Unknown
Knowledge
A24.8Supported
Multimodal
A13.4Preliminary
Long context
A33.3Preliminary

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

o3 · Reasoning: 10.9 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 · 100.0% direct comparison weight · independent of the model provider

Which benchmark families carry the weight?

MMLU-Pro · 8.2% direct comparison weight
Aime · 8.2% direct comparison weight
LiveCodeBench · 8.2% direct comparison weight
math500 · 8.1% direct comparison weight
IFBench · 8.0% direct comparison weight
Other families (8) · 59.3% 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 2026-10-23.

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 source	Rank before → after	Rank movement	Cohort remaining / lost support	Score after (change)
Artificial Analysis	Comparison support lost; no rank		203 remaining / 241 lost	Unavailable
ARC Prize	190 → 186	Up 4	444 remaining / 0 lost	13.8 (+2.9 points)
Datacurve	190 → 189	Up 1	444 remaining / 0 lost	8.8 (-2.1 points)
LiveCodeBench	190 → 189	Up 1	444 remaining / 0 lost	10.8 (-0.1 points)
Scale	190 → 190	No movement	444 remaining / 0 lost	12.5 (+1.6 points)
Terminal-Bench	190 → 190	No movement	444 remaining / 0 lost	10.8 (-0.1 points)
Cognition	190 → 192	Down 2	444 remaining / 0 lost	10.6 (-0.4 points)
LiveBench	189 → 187	Up 2	443 remaining / 1 lost	11.3 (+0.3 points)
SWE-rebench	190 → 188	Up 2	444 remaining / 0 lost	11.2 (+0.2 points)
Epoch AI	186 → 186	No movement	431 remaining / 13 lost	11.0 (+0.1 points)
Vals AI	190 → 189	Up 1	443 remaining / 1 lost	10.5 (-0.4 points)
Provider reports · Anthropic	190 → 190	No movement	444 remaining / 0 lost	10.9 (-0.0 points)
Provider reports · Moonshot AI	190 → 190	No movement	444 remaining / 0 lost	10.6 (-0.3 points)
Provider reports · Meta	190 → 190	No movement	444 remaining / 0 lost	10.9 (-0.0 points)
Provider reports · Mistral	190 → 190	No movement	444 remaining / 0 lost	10.9 (+0.0 points)
Provider reports · OpenAI	190 → 190	No movement	444 remaining / 0 lost	11.0 (+0.1 points)
All provider reports	190 → 190	No movement	444 remaining / 0 lost	10.7 (-0.2 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 evidence	Recomputed score	Change
arc-agi	13.4	+2.5
deepswe	8.8	-2.1
omniscience	9.3	-1.6
mmmu-pro	12.4	+1.4
hle	12.1	+1.2
aa-lcr	10.3	-0.6
epoch-game-puzzles	10.3	-0.6
ifbench	10.4	-0.5
simpleqa	11.4	+0.5
tau-bench	11.4	+0.5
mmlu-pro	11.3	+0.4
frontiercode	10.6	-0.4
critpt	11.2	+0.3
terminal-bench	10.7	-0.2
swe-rebench	11.2	+0.2
livebench-instructions	11.1	+0.2
multi-swe-bench	10.7	-0.2
gdp-pdf	11.1	+0.2
arc-agi-3	10.8	-0.2
livebench-language	11.1	+0.2
terminal-bench-science	11.1	+0.1
apex-agents	11.1	+0.1
enigma-eval	10.8	-0.1
harvey	10.8	-0.1
swe-atlas-qna	10.8	-0.1
livebench-coding	10.8	-0.1
itbench	10.8	-0.1
automationbench	10.8	-0.1
aime	11.0	+0.1
math500	10.8	-0.1
analyst-agent	11.0	+0.1
briefcase	10.9	-0.1
gpqa	11.0	+0.1
livebench-reasoning	10.9	-0.1
gmmlu	11.0	+0.1
vibe-code	10.9	-0.1
swe-atlas-test-writing	10.9	-0.0
vals-code-migration	10.9	-0.0
vals-finance-agent	10.9	-0.0
vals-legal-research	10.9	-0.0
vals-tax-agent	10.9	-0.0
frontiermath	10.9	+0.0
livecodebench	10.9	+0.0
vals-excel-modeling	10.9	-0.0
enterprise-ops	10.9	-0.0
livebench-math	10.9	+0.0
gdpval	10.9	+0.0
swe-atlas-refactoring	10.9	-0.0
livebench-data	10.9	-0.0
swe-bench-pro	10.9	+0.0

What evidence is missing from the fit?

Across all collected settings for this model: 22 observations, 17 contributing, 0 matched without graph weight and 5 excluded. Counts do not establish rank eligibility.

1 · Benchmark family or source protocol has not been reviewed
4 · No reviewed effort for this catalog observation; any separately collected effort measurement is counted separately

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 / source	Low	Medium	High	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 2026-10-18.	Not reported	Not reported	Not reported	88.33% Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
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.	Not reported	Not reported	Not reported	74.67% Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
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.	Not reported	Not reported	Not reported	1.14% Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
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.	Not reported	Not reported	Not reported	82.73% Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
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.	Not reported	Not reported	Not reported	20.05% Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
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.	Not reported	Not reported	Not reported	71.43% Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
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 2026-10-18.	Not reported	Not reported	Not reported	80.85% Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
MATH-500Artificial 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.	Not reported	Not reported	Not reported	99.2% Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
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 2026-10-18.	Not reported	Not reported	Not reported	85.35% Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
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.	Not reported	Not reported	Not reported	70.06% Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
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.	Not reported	Not reported	Not reported	-15.55 Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
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.	Not reported	Not reported	Not reported	38.57% Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
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.	Not reported	Not reported	Not reported	80.7% Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
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.	Not reported	Not reported	Not reported	37.12% Reported settings o3; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/o3; Mode from evaluator isReasoning metadata; effort budget not specified
ARC-AGI-2ARC Prize · v2 Semi-Private Benchmark details Original public board. Configuration and task coverage may differ; inspect source. Unit: percent. Source collected 2026-10-18.	1.99% Reported settings o3 (Low); Source model ID: o3-2025-04-16-low; 120 tasks	2.98% Reported settings o3 (Medium); Source model ID: o3-2025-04-16-medium; 120 tasks	6.53% Reported settings o3 (High); Source model ID: o3-2025-04-16-high; 120 tasks	Not reported
LiveCodeBenchLiveCodeBench · generation pass@1 Benchmark details Evaluation window 2024-08-01 to 2025-05-01; 454 shared tasks. Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	75.8% Reported settings O3 (High); Exact source model: o3__high; 454 tasks	Not 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 · 8,916 results across 259 models · Collected 2026-10-18

o3 · Reasoning

Compare with

One configuration, seven capabilities

Agentic

Hard reasoning

Coding

Human pref

Knowledge

Multimodal

Long context

Capability score evidence

Benchmark results