RankingEffort benchmarks

Claude Sonnet 4.6 effort benchmarks

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

Model

Claude Sonnet 4.6

107 measured results · 7 reported effort labels · 68 benchmark/harness combinations. Model evidence and sources →

Capability score evidence

Claude Sonnet 4.6 · Max: 23.2 overall comparison score. Meets the overall evidence requirements.

Which sources carry the weight?

Cognition · 21.6% direct comparison weight · independent
ARC Prize · 21.5% direct comparison weight · independent
Artificial Analysis · 21.5% direct comparison weight · independent
Vals AI · 18.9% direct comparison weight · independent
Epoch AI · 16.5% 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	21.7	-1.6
ARC Prize	21.3	-2.0
Datacurve	20.4	-2.8
LiveCodeBench	23.2	+0.0
Scale	24.6	+1.4
Terminal-Bench	23.2	-0.0
Cognition	27.3	+4.1
LiveBench	23.7	+0.5
SWE-rebench	23.1	-0.2
Epoch AI	29.0	+5.7
Vals AI	20.2	-3.1
Provider reports · Anthropic	23.2	-0.0
Provider reports · Moonshot AI	23.2	-0.0
Provider reports · Meta	23.2	-0.0
Provider reports · Mistral	23.2	+0.0
All provider reports	23.2	-0.0

Every benchmark family

Removed evidence	Recomputed score	Change
frontiercode	27.3	+4.1
deepswe	20.5	-2.8
arc-agi	21.2	-2.0
mmmu-pro	24.3	+1.1
hle	23.8	+0.6
simpleqa	23.8	+0.6
mmlu-pro	23.7	+0.5
aime	22.8	-0.5
vals-code-migration	22.8	-0.4
harvey	22.8	-0.4
terminal-bench	23.6	+0.4
epoch-game-puzzles	23.5	+0.3
multi-swe-bench	23.0	-0.3
livebench-instructions	23.5	+0.2
vals-legal-research	23.0	-0.2
aa-lcr	23.0	-0.2
vals-excel-modeling	23.0	-0.2
swe-atlas-qna	23.0	-0.2
livebench-language	23.4	+0.2
livebench-coding	23.1	-0.2
swe-rebench	23.1	-0.2
vibe-code	23.4	+0.1
frontiermath	23.1	-0.1
itbench	23.1	-0.1
livebench-reasoning	23.1	-0.1
scicode	23.3	+0.1
critpt	23.3	+0.1
gpqa	23.3	+0.1
swe-atlas-test-writing	23.1	-0.1
ifbench	23.3	+0.1
briefcase	23.1	-0.1
livebench-data	23.3	+0.1
automationbench	23.3	+0.1
omniscience	23.2	-0.1
arc-agi-3	23.2	-0.1
enigma-eval	23.3	+0.1
tau-bench	23.2	-0.0
vals-finance-agent	23.3	+0.0
apex-agents	23.3	+0.0
gdpval	23.2	-0.0
gmmlu	23.3	+0.0
livebench-math	23.3	+0.0
gdp-pdf	23.2	-0.0
swe-atlas-refactoring	23.2	-0.0
livecodebench	23.2	+0.0
math500	23.2	+0.0
swe-bench-pro	23.2	+0.0
analyst-agent	23.2	+0.0
enterprise-ops	23.2	-0.0

What evidence is missing from the fit?

Across all collected settings for this model: 151 observations, 80 contributing, 1 matched without graph weight and 70 excluded. Counts do not establish rank eligibility.

20 · Conflicting values for this configuration in the same evaluation
3 · Benchmark family or source protocol has not been reviewed
2 · Fewer than two distinct model identities remain in this evaluation
1 · Effort is not identified
17 · No reviewed effort for this catalog observation; any separately collected effort measurement is counted separately
1 · Microsoft report Table 11 and section 4.1: MAI Terminal-Bench removes timeouts and uses a minimal ReAct harness; comparator values are cited from other model releases. A common evaluation protocol is not established.
4 · No matched, identified effort unit; duplicates and unnamed comparators do not add evidence
22 · Supporting evidence outside the reviewed capability core

Inspect every observation and exclusion →

Capability profile

One configuration, seven capabilities

Comparison score · 0–100

AClaude Sonnet 4.6 · Max4 of 7 capabilities supported

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

Agentic
A40.8Supported
Hard reasoning
A26.2Supported
Coding
A28.3Supported
Human pref
Unknown
Knowledge
A19.2Supported
Multimodal
A19.1Preliminary
Long context
A62.6Preliminary

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 / source	Low	Medium	High	Max	Thinking 32K	Non-reasoning	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	20% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	Not reported	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.	Not reported	Not reported	Not reported	1,065.11 Elo Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	Not reported	Not reported
APEX-Agents-AAArtificial 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 reported	Not reported	Not reported	28.02% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	Not reported	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.	Not reported	Not reported	Not reported	80% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	68.33% Reported settings Claude Sonnet 4.6 (Non-reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6 69.33% Reported settings Claude Sonnet 4.6 (Non-reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-non-reasoning-low-effort	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.	Not reported	Not reported	Not reported	20.1% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	Not reported	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.	Not reported	Not reported	Not reported	3.14% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	0.86% Reported settings Claude Sonnet 4.6 (Non-reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6 0.86% Reported settings Claude Sonnet 4.6 (Non-reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-non-reasoning-low-effort	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.	Not reported	Not reported	Not reported	15.8% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	Not reported	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.	Not reported	Not reported	Not reported	1,295.41 Elo Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	Not reported	Not 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 reported	Not reported	Not reported	90.48% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	88.24% Reported settings Claude Sonnet 4.6 (Non-reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-non-reasoning-low-effort 89.24% Reported settings Claude Sonnet 4.6 (Non-reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6	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.	Not reported	Not reported	Not reported	87.47% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	79.9% Reported settings Claude Sonnet 4.6 (Non-reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6 79.7% Reported settings Claude Sonnet 4.6 (Non-reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-non-reasoning-low-effort	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	85.97% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	84.44% Reported settings Claude Sonnet 4.6 (Non-reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6	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.	Not reported	Not reported	Not reported	33.6% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	13.35% Reported settings Claude Sonnet 4.6 (Non-reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6 11.21% Reported settings Claude Sonnet 4.6 (Non-reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-non-reasoning-low-effort	Not reported
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 reported	Not reported	Not reported	56.6% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	41.16% Reported settings Claude Sonnet 4.6 (Non-reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6 42.38% Reported settings Claude Sonnet 4.6 (Non-reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-non-reasoning-low-effort	Not reported
ITBench-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Combined public and private Site Reliability Engineering (SRE) tasks · 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	39.8% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	Not reported	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.	Not reported	Not reported	Not reported	24.44% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	20.56% Reported settings Claude Sonnet 4.6 (Non-reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-non-reasoning-low-effort 20% Reported settings Claude Sonnet 4.6 (Non-reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6	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.	Not reported	Not reported	Not reported	73.29% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	70.58% Reported settings Claude Sonnet 4.6 (Non-reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6 69.19% Reported settings Claude Sonnet 4.6 (Non-reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-non-reasoning-low-effort	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.	Not reported	Not reported	Not reported	12.22 Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	-2.1 Reported settings Claude Sonnet 4.6 (Non-reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-non-reasoning-low-effort -3.55 Reported settings Claude Sonnet 4.6 (Non-reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6	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.	Not reported	Not reported	Not reported	40.87% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	36.52% Reported settings Claude Sonnet 4.6 (Non-reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-non-reasoning-low-effort 38.55% Reported settings Claude Sonnet 4.6 (Non-reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6	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.	Not reported	Not reported	Not reported	50.12% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	Not reported	Not reported
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 reported	Not reported	Not reported	75.73% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	79.53% Reported settings Claude Sonnet 4.6 (Non-reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6 78.95% Reported settings Claude Sonnet 4.6 (Non-reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-non-reasoning-low-effort	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.	Not reported	Not reported	Not reported	34.43% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	Not reported	Not 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 reported	Not reported	Not reported	53.03% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	42.42% Reported settings Claude Sonnet 4.6 (Non-reasoning, Low Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-non-reasoning-low-effort 46.21% Reported settings Claude Sonnet 4.6 (Non-reasoning, High Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6	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.	Not reported	Not reported	Not reported	71.16% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	Not reported	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.	Not reported	Not reported	Not reported	3.03% Reported settings Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort); Model source: https://artificialanalysis.ai/models/claude-sonnet-4-6-adaptive	Not reported	Not reported	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	60.42% Reported settings Claude Sonnet 4.6 (High); Source model ID: claude_sonnet_4_6_high; 120 tasks	58.33% Reported settings Claude Sonnet 4.6 (Max); Source model ID: claude_sonnet_4_6_max; 120 tasks	Not reported	Not reported	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.	Not reported	Not reported	29.9% Reported settings claude-sonnet-4-6; Reported effort: high; 113 unique tasks; 451 attempts; 4 runs	Not reported	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.	30.44% 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	37.49% 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	39.86% 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	40% 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 0.0013	Not reported	Not reported	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.	15.03% 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	21.1% 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	23.47% 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	24.31% 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 0.002	Not reported	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	76% Reported settings LiveBench official task evaluation; livebench-2026-06-25-AMPS_Hard; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	78.26% Reported settings LiveBench official task evaluation; livebench-2026-06-25-code_completion; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	80.28% Reported settings LiveBench official task evaluation; livebench-2026-06-25-code_generation; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	99.33% Reported settings LiveBench official task evaluation; livebench-2026-06-25-connections; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	92.68% Reported settings LiveBench official task evaluation; livebench-2026-06-25-consecutive_events; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	90% Reported settings LiveBench official task evaluation; livebench-2026-06-25-integrals_with_game; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	54.55% Reported settings LiveBench Mini-SWE-Agent, 250-step limit; livebench-2026-06-25-javascript; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	70% Reported settings LiveBench official task evaluation; livebench-2026-06-25-logic_with_navigation; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	94.12% Reported settings LiveBench official task evaluation; livebench-2026-06-25-math_comp; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	87.86% Reported settings LiveBench official task evaluation; livebench-2026-06-25-olympiad; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	59.4% Reported settings LiveBench official task evaluation; livebench-2026-06-25-paraphrase; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	56.98% Reported settings LiveBench official task evaluation; livebench-2026-06-25-plot_unscrambling; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	50% Reported settings LiveBench Mini-SWE-Agent, 250-step limit; livebench-2026-06-25-python; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	62.93% Reported settings LiveBench official task evaluation; livebench-2026-06-25-simplify; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	100% Reported settings LiveBench official task evaluation; livebench-2026-06-25-spatial; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	68.9% Reported settings LiveBench official task evaluation; livebench-2026-06-25-story_generation; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	61.65% Reported settings LiveBench official task evaluation; livebench-2026-06-25-summarize; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	45.08% Reported settings LiveBench official task evaluation; livebench-2026-06-25-tablejoin; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	96.08% Reported settings LiveBench official task evaluation; livebench-2026-06-25-tablereformat; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	73.08% Reported settings LiveBench official task evaluation; livebench-2026-06-25-theory_of_mind; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	23.33% Reported settings LiveBench Mini-SWE-Agent, 250-step limit; livebench-2026-06-25-typescript; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	72% Reported settings LiveBench official task evaluation; livebench-2026-06-25-typos; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	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	96% Reported settings LiveBench official task evaluation; livebench-2026-06-25-zebra_puzzle; claude-sonnet-4-6-thinking-auto-medium-effort; checkpoint=claude-sonnet-4-6	Not reported	Not reported	Not reported	Not reported	Not reported
Epoch · Chess Puzzles · v1.0.2Epoch AI · Epoch AI Inspect; task 1.0.2; 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.	Not reported	Not reported	Not reported	Not reported	13% Reported settings Epoch AI Inspect; task 1.0.2; scorer model_extracted_exact_match; model=claude-sonnet-4-6_32K; run=N5beL4W3WahAubJgDKivnE; mean accuracy; https://epoch.ai/data/benchmarks.csv	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.	Not reported	8% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=claude-sonnet-4-6_medium; run=eJBzZyT5ohb8ieaJCum5rK; mean accuracy; https://epoch.ai/data/benchmarks.csv	5% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=claude-sonnet-4-6_high; run=5Q9RNC2JfEmmvNFFE5CQCf; mean accuracy; https://epoch.ai/data/benchmarks.csv	3% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=claude-sonnet-4-6_max; run=TDT98jZXHbUMFYTphGEpqC; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported	Not reported	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.	Not reported	83.33% Reported settings Epoch AI Inspect; task 1.0.11; scorer choice; model=claude-sonnet-4-6_medium; run=e8MXnAM8HMF8eEzYtYA7xv; mean accuracy; https://epoch.ai/data/benchmarks.csv	83.33% Reported settings Epoch AI Inspect; task 1.0.11; scorer choice; model=claude-sonnet-4-6_high; run=DmdKPD9Z9RXwYyATAoG4Uw; mean accuracy; https://epoch.ai/data/benchmarks.csv	78.79% Reported settings Epoch AI Inspect; task 1.0.11; scorer choice; model=claude-sonnet-4-6_max; run=SEePRKRcv8oqn5HpD2KFxB; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported	Not reported	Not reported
Epoch · GPQA diamond · v1.0.6Epoch AI · Epoch AI Inspect; task 1.0.6; 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.	Not reported	Not reported	Not reported	Not reported	87.37% Reported settings Epoch AI Inspect; task 1.0.6; scorer choice; model=claude-sonnet-4-6_32K; run=58eQmyCfPa3FufhXJFvAL2; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported	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.	16% Reported settings Epoch AI Inspect; task 1.0.4; scorer exact_match_next_move_scorer; model=claude-sonnet-4-6_low; run=YdyvoMQ4mA5V8Ymg6Tksxu; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported	Not reported	Not reported	Not reported	Not reported	14% Reported settings Epoch AI Inspect; task 1.0.4; scorer exact_match_next_move_scorer; model=claude-sonnet-4-6; run=VjfWxdhdNerrh4huSHT8PT; mean accuracy; https://epoch.ai/data/benchmarks.csv
Epoch · OTIS Mock AIME 2024-2025 · v1.0.5Epoch AI · Epoch AI Inspect; task 1.0.5; 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.	Not reported	Not reported	Not reported	Not reported	85.79% Reported settings Epoch AI Inspect; task 1.0.5; scorer model_extracted_exact_match; model=claude-sonnet-4-6_32K; run=9LWdzVN5w4ihNvncsaCWMu; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported	Not reported
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.	Not reported	82.22% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=claude-sonnet-4-6_medium; run=kDNNbUYXhrAH5uDBGykFaf; mean accuracy; https://epoch.ai/data/benchmarks.csv	75.56% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=claude-sonnet-4-6_high; run=oXZ7ZVZtADrwFPKAe6zTvL; mean accuracy; https://epoch.ai/data/benchmarks.csv	71.11% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=claude-sonnet-4-6_max; run=XnQVH5qSfxVjpzoPwUEmnh; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported	Not reported	Not reported
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	35.5% Reported settings Epoch AI Inspect; task 1.0.0; scorer simpleqa_scorer; model=claude-sonnet-4-6_high; run=LJyKf8Xy4dymxNHbGLwNdA; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported	Not reported	Not reported	Not reported
Epoch · SimpleQA Verified · v1.2.0Epoch AI · Epoch AI Inspect; task 1.2.0; 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	32.8% Reported settings Epoch AI Inspect; task 1.2.0; scorer simpleqa_scorer; model=claude-sonnet-4-6_max; run=Zs5fCvNzaz8LhqgwDSAztU; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported	Not reported	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	51.04% Reported settings {"checkpoint":"claude-sonnet-4-6","evaluatorModel":"anthropic/claude-sonnet-4-6","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":"max","reasoning":true,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null}	Not reported	Not reported	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	39.89% Reported settings {"checkpoint":"claude-sonnet-4-6","evaluatorModel":"anthropic/claude-sonnet-4-6","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":"max","reasoning":true,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null}	Not reported	Not reported	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	38.46% Reported settings {"checkpoint":"claude-sonnet-4-6","evaluatorModel":"anthropic/claude-sonnet-4-6","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":"max","reasoning":true,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null}	Not reported	Not reported	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	60.15% Reported settings {"checkpoint":"claude-sonnet-4-6","evaluatorModel":"anthropic/claude-sonnet-4-6","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":"max","reasoning":true,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null}	Not reported	Not reported	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	87.34% Reported settings {"checkpoint":"claude-sonnet-4-6","evaluatorModel":"anthropic/claude-sonnet-4-6","sourceUpdatedOn":"2026-09-01","reasoning_effort":null,"compute_effort":"max","reasoning":true,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null}	Not reported	Not reported	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	Not reported	57.3% Reported settings {"checkpoint":"claude-sonnet-4-6","evaluatorModel":"anthropic/claude-sonnet-4-6","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":"max","reasoning":false,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null}	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	51.48% Reported settings {"checkpoint":"claude-sonnet-4-6","evaluatorModel":"anthropic/claude-sonnet-4-6","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":"max","reasoning":true,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":"OpenHands"}	Not reported	Not reported	Not reported
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	5% Reported settings {"checkpoint":"claude-sonnet-4-6","evaluatorModel":"anthropic/claude-sonnet-4-6","sourceUpdatedOn":"2026-09-10","reasoning_effort":null,"compute_effort":"max","reasoning":true,"temperature":1,"top_p":null,"max_output_tokens":128000,"verbosity":null,"provider":"Anthropic","harness":null}	Not reported	Not reported	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 Sonnet 4.6 effort benchmarks

Claude Sonnet 4.6

Capability score evidence

One configuration, seven capabilities

Agentic

Hard reasoning

Coding

Human pref

Knowledge

Multimodal

Long context