RankingEffort benchmarks

Claude Opus 5 effort benchmarks

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

Model

Claude Opus 5

153 measured results · 6 reported effort labels · 68 benchmark/harness combinations. Model evidence and sources →

Capability score evidence

Claude Opus 5 · Medium: 75.7 overall comparison score. Meets the overall evidence requirements.

Which sources carry the weight?

Cognition · 46.1% direct comparison weight · independent
Artificial Analysis · 34.2% direct comparison weight · independent
Datacurve · 19.7% 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 evidence	Recomputed score	Change
Artificial Analysis	81.6	+5.9
ARC Prize	76.6	+0.9
Datacurve	75.3	-0.3
LiveCodeBench	75.6	-0.0
Scale	76.4	+0.7
Terminal-Bench	75.6	-0.1
Cognition	64.0	-11.7
LiveBench	75.9	+0.3
SWE-rebench	75.5	-0.2
Epoch AI	77.2	+1.5
Vals AI	75.2	-0.5
Provider reports · Anthropic	75.6	-0.1
Provider reports · Moonshot AI	75.6	-0.1
Provider reports · Meta	75.6	-0.0
Provider reports · Mistral	75.7	+0.0
All provider reports	75.5	-0.1

Every benchmark family

Removed evidence	Recomputed score	Change
frontiercode	64.0	-11.7
arc-agi	76.9	+1.2
mmmu-pro	76.4	+0.8
scicode	76.1	+0.4
ifbench	75.3	-0.4
gpqa	76.0	+0.3
aime	75.3	-0.3
deepswe	75.3	-0.3
hle	75.9	+0.3
swe-atlas-qna	75.4	-0.2
arc-agi-3	75.5	-0.2
livebench-instructions	75.9	+0.2
gdp-pdf	75.8	+0.2
tau-bench	75.5	-0.2
swe-atlas-test-writing	75.5	-0.2
swe-rebench	75.5	-0.2
simpleqa	75.8	+0.2
livebench-coding	75.5	-0.2
multi-swe-bench	75.5	-0.1
harvey	75.5	-0.1
livebench-language	75.8	+0.1
mmlu-pro	75.8	+0.1
automationbench	75.8	+0.1
omniscience	75.8	+0.1
livecodebench	75.5	-0.1
swe-atlas-refactoring	75.6	-0.1
livebench-reasoning	75.6	-0.1
terminal-bench	75.6	-0.1
frontiermath	75.6	-0.1
aa-lcr	75.6	-0.1
livebench-data	75.7	+0.1
epoch-game-puzzles	75.7	+0.1
critpt	75.7	+0.1
gdpval	75.7	+0.1
vibe-code	75.6	-0.1
vals-legal-research	75.6	-0.1
enterprise-ops	75.6	-0.1
enigma-eval	75.6	-0.1
math500	75.6	-0.1
gmmlu	75.6	-0.0
briefcase	75.6	-0.0
itbench	75.6	-0.0
apex-agents	75.7	+0.0
vals-excel-modeling	75.6	-0.0
vals-code-migration	75.6	-0.0
analyst-agent	75.6	-0.0
livebench-math	75.6	-0.0
vals-finance-agent	75.6	-0.0
swe-bench-pro	75.7	+0.0

What evidence is missing from the fit?

Across all collected settings for this model: 259 observations, 144 contributing, 2 matched without graph weight and 113 excluded. Counts do not establish rank eligibility.

5 · Benchmark family or source protocol has not been reviewed
3 · Evaluator has not been reviewed for independent admission
5 · Effort is not identified
1 · One run-task used an Opus 4.8 refusal fallback; mixed checkpoint result excluded from ranking
1 · Source discloses an Opus 4.8 refusal fallback; mixed checkpoint result excluded from ranking
38 · No reviewed effort for this catalog observation; any separately collected effort measurement is counted separately
46 · Supporting evidence outside the reviewed capability core
1 · Secondary GDPval-AA v2 leaderboard report; the original evaluator already supplies this exact configuration. Reprinted or rerated board snapshots are not additional evaluation evidence.
3 · No matched opponent in this evaluation unit
1 · Metric turns is outside the task-success core
1 · Mixed source task revisions and best-of-three versus provider reporting prevent a uniform common-core unit.
1 · Different native coding harnesses are mixed; exact shared agent configuration is not established.
7 · Legacy self-report lacks a reviewed comparison configuration

Inspect every observation and exclusion →

Capability profile

One configuration, seven capabilities

Comparison score · 0–100

AClaude Opus 5 · Medium3 of 7 capabilities supported

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

Agentic
A60.3Supported
Hard reasoning
A54.8Supported
Coding
A78.3Supported
Human pref
Unknown
Knowledge
A58.7Preliminary
Multimodal
A52.0Preliminary
Long context
A74.7Preliminary

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 / source	Low	Medium	High	XHigh	Max	Unspecified
AA-AnalystAgentArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of tasks solved on all 5 of 5 attempts · 80 tasks, 5 attempts per task · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	53.75% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
AA-Briefcase EloArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details AA-Briefcase 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 2026-09-12.	1,215.73 Elo Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	1,439.04 Elo Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	1,556.66 Elo Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	1,624.56 Elo Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	1,644.9 Elo Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
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.	81.33% Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	82% Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	79% Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	80.33% Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	79.33% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
AutomationBench-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of task objectives completed with no guardrail violations · Higher is better · Benchmark developed by Zapier · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	51.79% Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	54.34% Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	53.56% Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	53.22% Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	56.57% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
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.	23.14% Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	26.86% Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	28.29% Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	27.71% Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	29.14% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
EnterpriseOps-Gym-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Overall task success rate across all eight enterprise domains (oracle tool mode) · Benchmark developed by ServiceNow Research · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	47.48% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
GDP.pdf: All-passArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of attempts where every atomic criterion passed · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	17.2% Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	20% Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	19.6% Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	21% Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	21.6% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
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.	1,371.52 Elo Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	1,524.87 Elo Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	1,628.94 Elo Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	1,708.14 Elo Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	1,735.09 Elo Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not 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.	88.89% Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	91.92% Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	93.74% Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	93.74% Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	93.23% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
Harvey LAB-AA: Criterion Pass RateArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of atomic pass/fail rubric criteria the deliverables satisfy, graded by a single LLM judge · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	93.46% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
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.	43.42% Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	51.3% Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	52.83% Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	54.4% Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	54.87% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
MLCR-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of tasks judged accurate, complete, and concise · Expert + compound (hard) sets · Benchmark developed by Wisedocs · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	53.89% Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	56.11% Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	59.44% Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	58.33% Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	55.56% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
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.	79.83% Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	81.62% Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	82.43% Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	83.99% Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	84.74% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
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.	28.55 Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	31.02 Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	33.72 Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	35.38 Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	37.07 Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
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.	55.95% Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	57.07% Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	58.88% Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	59.5% Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	60.87% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
SciCodeArtificial 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.	49.19% Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	51.5% Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	55.44% Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	55.67% Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	56.37% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
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.	30.31% Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	38.56% Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	44.74% Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	43.3% Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	42.06% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
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.	76.4% Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	86.14% Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	87.64% Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	88.01% Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	89.14% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not reported
Terminal-Bench v4.0Artificial Analysis · mini-SWE-agent v2.4.6; 66 tasks; pass@1 averaged over 3 repeats; 500 steps; task timeout up to 8 hours Benchmark details Benchmark developed by the Laude Institute, Stanford, and open-source contributors · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	26.26% Reported settings Claude Opus 5 (Adaptive Reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-low	34.34% Reported settings Claude Opus 5 (Adaptive Reasoning, Medium Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-medium	45.96% Reported settings Claude Opus 5 (Adaptive Reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-high	46.46% Reported settings Claude Opus 5 (Adaptive Reasoning, Xhigh Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5-xhigh	48.99% Reported settings Claude Opus 5 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-opus-5	Not 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.	Not reported	Not reported	88.33% Reported settings Claude Opus 5 (High); Source model ID: anthropic-claude-opus-5-high; 120 tasks	Not reported	90.42% Reported settings Claude Opus 5 (Max); Source model ID: anthropic-claude-opus-5-max; 120 tasks	Not reported
DeepSWE v1.1Datacurve · mini-swe-agent Benchmark details Original public board. Configuration and task coverage may differ; inspect source. Unit: percent. Source collected 2026-09-12.	58.1% Reported settings claude-opus-5; Reported effort: low; 113 unique tasks; 449 attempts; 4 runs	68.9% Reported settings claude-opus-5; Reported effort: medium; 113 unique tasks; 447 attempts; 4 runs	72.8% Reported settings claude-opus-5; Reported effort: high; 113 unique tasks; 449 attempts; 4 runs	73.2% Reported settings claude-opus-5; Reported effort: xhigh; 113 unique tasks; 447 attempts; 4 runs	73.6% Reported settings claude-opus-5; Reported effort: max; 113 unique tasks; 444 attempts; 4 runs	Not reported
VulcanBench v3 · Report 10VulcanBench · VulcanBench bare-bones API loop, deterministic hidden tests in Docker Benchmark details Frozen v3, 23 post-cutoff repository tasks. Published pass@1; repeat counts and time budgets vary by report. Display only pending comparability review. Unit: percent. Source collected 2026-09-12.	87% Reported settings Claude Opus 5; source label: low; low effort; 23 tasks; evaluation 2026-07-26. See source for repeat counts and wall-clock budget.	82.6% Reported settings Claude Opus 5; source label: medium; medium effort; 23 tasks; evaluation 2026-07-26. See source for repeat counts and wall-clock budget.	78.3% Reported settings Claude Opus 5; source label: high; high effort; 23 tasks; evaluation 2026-07-26. See source for repeat counts and wall-clock budget.	Not reported	Not reported	Not reported
ARC-AGI-3 · standardARC Prize · Standard Benchmark details ARC-AGI-3; Standard. Open each result for its exact effort, task subset, metric and protocol details. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	30.16% Reported settings ARC-AGI-3 Semi-Private verified summary; Standard harness; source group anthropic-claude-opus-5; variant high; public-demo task scores are excluded	Not reported	Not reported	Not reported
FrontierCode 1.1 · extended-claude-codeCognition · claude-code Benchmark details FrontierCode 1.1; claude-code. Open each result for its exact effort, task subset, metric and protocol details. Unit: percent. Source collected 2026-09-12.	55.75% Reported settings FrontierCode 1.1; extended; 150 tasks; claude-code; source effort low; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0.0041	63.63% Reported settings FrontierCode 1.1; extended; 150 tasks; claude-code; source effort medium; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0.004	58.45% Reported settings FrontierCode 1.1; extended; 150 tasks; claude-code; source effort high; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0.0013	56.87% Reported settings FrontierCode 1.1; extended; 150 tasks; claude-code; source effort xhigh; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0.0097	58.94% Reported settings FrontierCode 1.1; extended; 150 tasks; claude-code; source effort max; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0.0067	Not reported
FrontierCode 1.1 · main-claude-codeCognition · claude-code Benchmark details FrontierCode 1.1; claude-code. Open each result for its exact effort, task subset, metric and protocol details. Unit: percent. Source collected 2026-09-12.	41.95% Reported settings FrontierCode 1.1; main; 100 tasks; claude-code; source effort low; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0.0021	53.38% Reported settings FrontierCode 1.1; main; 100 tasks; claude-code; source effort medium; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0.006	47.99% Reported settings FrontierCode 1.1; main; 100 tasks; claude-code; source effort high; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0.002	43.65% Reported settings FrontierCode 1.1; main; 100 tasks; claude-code; source effort xhigh; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0.0144	48.04% Reported settings FrontierCode 1.1; main; 100 tasks; claude-code; source effort max; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0.0101	Not reported
Terminal-Bench 4.0 · claudecodeTerminal-Bench · Claude Code Benchmark details Terminal-Bench 4.0; Claude Code. Open each result for its exact effort, task subset, metric and protocol details. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	51.8% Reported settings Terminal-Bench 4.0.0; 66 tasks; Claude Code; effort max; reported margin ±3.4 percentage points, confidence level unspecified; release date 2026-07-24; board checked 2026-09-07	Not reported
SWE Atlas · QnA · claude codeScale · claude code Benchmark details 124 code comprehension tasks; strict all-rubrics pass metric; Opus 4.5 judge; source-reported agent harness. Repositories drawn from SWE-bench Pro; distinct tasks. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	63.17% Reported settings 124 code comprehension tasks; strict all-rubrics pass metric; Opus 4.5 judge; source-reported agent harness. Repositories drawn from SWE-bench Pro; distinct tasks.; published row 2026-08-04T16:11:53.000Z; reported margin ±5.01 percentage points; No contamination annotation reported (not proof of absence)	Not reported	Not reported
SWE Atlas · Test Writing · claude codeScale · claude code Benchmark details 90 test-writing tasks; execution, mutation checks and rubric judging; source-reported agent harness. Repositories drawn from SWE-bench Pro; distinct tasks. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	62.22% Reported settings 90 test-writing tasks; execution, mutation checks and rubric judging; source-reported agent harness. Repositories drawn from SWE-bench Pro; distinct tasks.; published row 2026-08-04T16:09:52.000Z; reported margin ±5.58 percentage points; No contamination annotation reported (not proof of absence)	Not reported	Not reported
LiveBench · AMPS Hard · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	99.01% Reported settings LiveBench official task evaluation; livebench-2026-06-25-AMPS_Hard; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · code completion · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	82.61% Reported settings LiveBench official task evaluation; livebench-2026-06-25-code_completion; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · code generation · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	80.28% Reported settings LiveBench official task evaluation; livebench-2026-06-25-code_generation; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · connections · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	99.33% Reported settings LiveBench official task evaluation; livebench-2026-06-25-connections; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · consecutive events · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	77.57% Reported settings LiveBench official task evaluation; livebench-2026-06-25-consecutive_events; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · integrals with game · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	97% Reported settings LiveBench official task evaluation; livebench-2026-06-25-integrals_with_game; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · javascript · 2026-06-25LiveBench · LiveBench Mini-SWE-Agent, 250-step limit Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	77.27% Reported settings LiveBench Mini-SWE-Agent, 250-step limit; livebench-2026-06-25-javascript; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · logic with navigation · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	86% Reported settings LiveBench official task evaluation; livebench-2026-06-25-logic_with_navigation; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · math comp · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	94.12% Reported settings LiveBench official task evaluation; livebench-2026-06-25-math_comp; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · olympiad · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	92.8% Reported settings LiveBench official task evaluation; livebench-2026-06-25-olympiad; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · paraphrase · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	65.73% Reported settings LiveBench official task evaluation; livebench-2026-06-25-paraphrase; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · plot unscrambling · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	74.73% Reported settings LiveBench official task evaluation; livebench-2026-06-25-plot_unscrambling; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · python · 2026-06-25LiveBench · LiveBench Mini-SWE-Agent, 250-step limit Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	75% Reported settings LiveBench Mini-SWE-Agent, 250-step limit; livebench-2026-06-25-python; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · simplify · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	61.58% Reported settings LiveBench official task evaluation; livebench-2026-06-25-simplify; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · spatial · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	100% Reported settings LiveBench official task evaluation; livebench-2026-06-25-spatial; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · story generation · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	61.32% Reported settings LiveBench official task evaluation; livebench-2026-06-25-story_generation; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · summarize · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	66.43% Reported settings LiveBench official task evaluation; livebench-2026-06-25-summarize; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · tablejoin · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	51.96% Reported settings LiveBench official task evaluation; livebench-2026-06-25-tablejoin; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · tablereformat · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	94.12% Reported settings LiveBench official task evaluation; livebench-2026-06-25-tablereformat; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · theory of mind · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	78.85% Reported settings LiveBench official task evaluation; livebench-2026-06-25-theory_of_mind; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · typescript · 2026-06-25LiveBench · LiveBench Mini-SWE-Agent, 250-step limit Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	43.33% Reported settings LiveBench Mini-SWE-Agent, 250-step limit; livebench-2026-06-25-typescript; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · typos · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	92% Reported settings LiveBench official task evaluation; livebench-2026-06-25-typos; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
LiveBench · zebra puzzle · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	100% Reported settings LiveBench official task evaluation; livebench-2026-06-25-zebra_puzzle; claude-opus-5-max-effort; checkpoint=claude-opus-5	Not reported
SWE-rebench · 2026-06 · 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 2026-09-12.	Not reported	Not reported	60% Reported settings SWE-rebench standardized ReAct tools; swe-rebench-2026-06-tools; Opus 5 [high]; checkpoint=claude-opus-5	Not reported	Not reported	Not reported
Epoch · Chess Puzzles · v1.1.6Epoch AI · Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. Unit: percent. Source collected 2026-09-12.	20% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=claude-opus-5_low; run=3dXpsUjaVEbBN72dnw4Vzq; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported	Not reported	Not reported	42% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=claude-opus-5_max; run=BiaKGj9CUTKhhNwW9SM4qt; mean accuracy; https://epoch.ai/data/benchmarks.csv	33% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=claude-opus-5; run=UxakyoDNbLKtxnxeMsa7cU; mean accuracy; https://epoch.ai/data/benchmarks.csv
Epoch · FrontierMath Tier 4 · v2.0.0Epoch AI · Epoch AI Inspect; task 2.0.0; scorer verification_code Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. FrontierMath was developed with OpenAI funding and unequal access to part of the question set; this result is from Epoch's evaluation. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	73.17% Reported settings Epoch AI Inspect; task 2.0.0; scorer verification_code; model=claude-opus-5_max; run=ji3kMn49Jcr9t4whxcpjj4; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported
Epoch · FrontierMath Tiers 1–3 · v2.0.0Epoch AI · Epoch AI Inspect; task 2.0.0; scorer verification_code Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. FrontierMath was developed with OpenAI funding and unequal access to part of the question set; this result is from Epoch's evaluation. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	85.61% Reported settings Epoch AI Inspect; task 2.0.0; scorer verification_code; model=claude-opus-5_max; run=ff7GeS7rdperjxSBZ7MHZY; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported
Epoch · GPQA diamond · v1.0.11Epoch AI · Epoch AI Inspect; task 1.0.11; scorer choice Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. Unit: percent. Source collected 2026-09-12.	87.88% Reported settings Epoch AI Inspect; task 1.0.11; scorer choice; model=claude-opus-5_low; run=WfaJS4L7Xdz5ZBXgJMbRGb; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported	Not reported	Not reported	93.88% Reported settings Epoch AI Inspect; task 1.0.11; scorer choice; model=claude-opus-5_max; run=D3mtasTumfvjR4fjmnHWRy; mean accuracy; https://epoch.ai/data/benchmarks.csv	92.93% Reported settings Epoch AI Inspect; task 1.0.11; scorer choice; model=claude-opus-5; run=mLmUnBuX6DYKtb6x46GteM; mean accuracy; https://epoch.ai/data/benchmarks.csv
Epoch · Mystery Game Puzzles · v1.0.4 · Tool submissionEpoch AI · Epoch AI Inspect; task 1.0.4; scorer agent_submitted_next_move_scorer Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	59% Reported settings Epoch AI Inspect; task 1.0.4; scorer agent_submitted_next_move_scorer; model=claude-opus-5_max; run=FfAvNS5H9bMDSBF4JNNiDz; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported
Epoch · Mystery Game Puzzles · v1.0.4 · Exact moveEpoch AI · Epoch AI Inspect; task 1.0.4; scorer exact_match_next_move_scorer Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	Not reported	37% Reported settings Epoch AI Inspect; task 1.0.4; scorer exact_match_next_move_scorer; model=claude-opus-5; run=bWnggCv6to2QKSjngrtuos; mean accuracy; https://epoch.ai/data/benchmarks.csv
Epoch · OTIS Mock AIME 2024-2025 · v1.1.6Epoch AI · Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. Unit: percent. Source collected 2026-09-12.	93.33% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=claude-opus-5_low; run=Z5FAHm8aqmk8NUUm7tojng; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported	Not reported	Not reported	98.89% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=claude-opus-5_max; run=KxvHAT9SV2dTDn7jBnTk6f; mean accuracy; https://epoch.ai/data/benchmarks.csv	97.78% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=claude-opus-5; run=gsBidnvEcmCfLnn5sxePtk; mean accuracy; https://epoch.ai/data/benchmarks.csv
Epoch · SimpleQA Verified · v1.0.0Epoch AI · Epoch AI Inspect; task 1.0.0; scorer simpleqa_scorer Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	59.9% Reported settings Epoch AI Inspect; task 1.0.0; scorer simpleqa_scorer; model=claude-opus-5_max; run=Zftz2yk4X5NF28JQzjN3hZ; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported
Vals AI · Finance Agent v2Vals AI · Vals FAB v2 shared six-tool harness; two-hour task limit Benchmark details Overall weighted rubric accuracy, averaged over three runs on 450 tasks. Dealbreaker failures and timeouts score zero. Related categories and all-pass scores are not additional votes. Three-model judge panel; private test set. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	58.63% Reported settings {"checkpoint":"claude-opus-5","evaluatorModel":"anthropic/claude-opus-5","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":"max","reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null} One run-task used an Opus 4.8 refusal fallback; mixed checkpoint result excluded from ranking Source label needs review	Not reported
Vals AI · Code Migration v1Vals AI · Vals Code Migration shared agent harness Benchmark details Hidden behavior-test pass rate, equally weighted across 40 repositories (30 CLI and 10 COBOL-to-Java). Anti-cheat failures score zero. Code-quality ratings and language splits are not additional votes. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	57.47% Reported settings {"checkpoint":"claude-opus-5","evaluatorModel":"anthropic/claude-opus-5","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":"max","reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null}	Not reported
Vals AI · Legal Research Bench v1Vals AI · Vals Legal Research shared five-tool harness; three-hour task limit Benchmark details All-pass accuracy on 208 private legal research tasks. Every expert rubric check must pass; GPT-5.4 judges responses. Partial-credit scores and practice-area splits are not additional votes. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	55.29% Reported settings {"checkpoint":"claude-opus-5","evaluatorModel":"anthropic/claude-opus-5","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":"max","reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null}	Not reported
Vals AI · Excel Modeling Benchmark v1Vals AI · Vals EMB shared workbook agent harness; 3.5-hour task limit Benchmark details Mean of seven equally weighted task categories. Template submissions use recalculated Excel cell accuracy; scratch submissions use rubric checks judged by GPT-5.4 Mini XHigh. Partial credit does not mean task completion. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	73.56% Reported settings {"checkpoint":"claude-opus-5","evaluatorModel":"anthropic/claude-opus-5","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":"max","reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null}	Not reported
Vals AI · MMLU-Pro v1Vals AI · Vals MMLU-Pro five-shot chain-of-thought; regex answer grading Benchmark details Mean accuracy across 14 subjects under Vals five-shot prompting. Subject splits share one family and are not additional votes. Vals labels this benchmark archived; it is historical corroboration, not ongoing frontier evaluation. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	91.59% Reported settings {"checkpoint":"claude-opus-5","evaluatorModel":"anthropic/claude-opus-5","sourceUpdatedOn":"2026-09-01","reasoning_effort":null,"compute_effort":"max","reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null}	Not reported
Vals AI · Terminal-Bench 2.1Vals AI · Vals Terminus 2 harness; pass@1 on 89 tasks Benchmark details Overall task pass rate under the shared Terminus 2 configuration. All task tests must pass. Difficulty splits share one family. Fable 5 and Opus 5 mixed-checkpoint fallback runs are excluded. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	84.64% Reported settings {"checkpoint":"claude-opus-5","evaluatorModel":"anthropic/claude-opus-5","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":"high","reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null} Source discloses an Opus 4.8 refusal fallback; mixed checkpoint result excluded from ranking Source label needs review	Not reported	Not reported	Not reported
Vals AI · Vibe Code Bench v1.1Vals AI · Vals modified OpenHands harness; five hours or 1,000 turns per app Benchmark details Mean app test pass rate on the canonical v1.1 test set. Browser Use executes workflow tests; each test requires at least 90% of its substeps. Automated judging can introduce error. Alternate agents are excluded. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	Not reported	88.4% Reported settings {"checkpoint":"claude-opus-5","evaluatorModel":"anthropic/claude-opus-5","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":null,"reasoning":null,"temperature":null,"top_p":null,"max_output_tokens":null,"verbosity":null,"provider":"Anthropic","harness":"OpenHands"}
Vals AI · Harvey Legal Agent Benchmark v1Vals AI · Vals Harvey held-out legal file-work protocol; internet disabled Benchmark details Mean task resolution across two judges (GPT-5.5 Medium and Sonnet 4.6), requiring every criterion to pass. Criteria pass rates and practice-area splits do not add votes. Shared Harvey family; Fable 5 fallback run excluded. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	6.67% Reported settings {"checkpoint":"claude-opus-5","evaluatorModel":"anthropic/claude-opus-5","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":"max","reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null}	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 · 7,424 results across 240 models · Collected 2026-09-12

Claude Opus 5 effort benchmarks

Claude Opus 5

Capability score evidence

One configuration, seven capabilities

Agentic

Hard reasoning

Coding

Human pref

Knowledge

Multimodal

Long context