RankingEffort benchmarks

Claude Opus 4.8

Compare models

95 measured results · 6 effort settings · 76 benchmark/harness combinations

Model evidence and sources →

Change model

Model

Current configurationSources

Compare with

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

Compare withModel and configuration

Capability profile

One configuration, seven capabilities

Comparison score · 0–100

AClaude Opus 4.8 · Max4 of 7 capabilities supported

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

Agentic
A57.7Supported
Hard reasoning
A58.2Supported
Coding
A55.7Supported
Human pref
Unknown
Knowledge
A48.4Supported
Multimodal
Unknown
Long context
A63.6Preliminary

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

Capability score evidence

Claude Opus 4.8 · Max: 52.9 overall comparison score. Meets the overall evidence requirements.

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

Which sources carry the weight?

Moonshot AI · 24.7% direct comparison weight · provider report
Artificial Analysis · 13.3% direct comparison weight · independent of the model provider
Cognition · 13.2% direct comparison weight · independent of the model provider
Terminal-Bench · 12.9% direct comparison weight · independent of the model provider
Vals AI · 12.6% direct comparison weight · independent of the model provider
LiveBench · 12.2% direct comparison weight · independent of the model provider
Epoch AI · 11.0% direct comparison weight · independent of the model provider

Which benchmark families carry the weight?

Terminal-Bench · 15.2% direct comparison weight
FrontierCode · 13.2% direct comparison weight
GPQA · 3.2% direct comparison weight
AutomationBench · 3.2% direct comparison weight
Humanity’s Last Exam · 3.2% direct comparison weight
Other families (37) · 61.9% direct comparison weight

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

How sensitive is this score?

The audit removes one source at a time and refits against the same overall reference panel. These published results use the matching dataset audited 2026-10-23.

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

What changes when a source is removed?
Removed source	Rank before → after	Rank movement	Cohort remaining / lost support	Score after (change)
Artificial Analysis	62 → 62	No movement	203 remaining / 241 lost	53.9 (+1.0 points)
ARC Prize	65 → 64	Up 1	444 remaining / 0 lost	56.5 (+3.6 points)
Datacurve	65 → 66	Down 1	444 remaining / 0 lost	47.7 (-5.3 points)
LiveCodeBench	65 → 65	No movement	444 remaining / 0 lost	52.9 (-0.0 points)
Scale	65 → 66	Down 1	444 remaining / 0 lost	54.2 (+1.3 points)
Terminal-Bench	65 → 59	Up 6	444 remaining / 0 lost	57.0 (+4.1 points)
Cognition	65 → 63	Up 2	444 remaining / 0 lost	50.9 (-2.1 points)
LiveBench	65 → 65	No movement	443 remaining / 1 lost	54.0 (+1.1 points)
SWE-rebench	65 → 66	Down 1	444 remaining / 0 lost	52.3 (-0.7 points)
Epoch AI	65 → 65	No movement	431 remaining / 13 lost	54.2 (+1.3 points)
Vals AI	65 → 67	Down 2	443 remaining / 1 lost	49.9 (-3.1 points)
Provider reports · Anthropic	65 → 65	No movement	444 remaining / 0 lost	53.0 (+0.1 points)
Provider reports · Moonshot AI	65 → 67	Down 2	444 remaining / 0 lost	51.1 (-1.8 points)
Provider reports · Meta	65 → 65	No movement	444 remaining / 0 lost	52.9 (-0.1 points)
Provider reports · Mistral	65 → 65	No movement	444 remaining / 0 lost	52.9 (+0.0 points)
Provider reports · OpenAI	65 → 65	No movement	444 remaining / 0 lost	52.9 (-0.0 points)
All provider reports	65 → 67	Down 2	444 remaining / 0 lost	51.2 (-1.7 points)

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

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

Every benchmark family

Removed evidence	Recomputed score	Change
terminal-bench	57.7	+4.7
deepswe	48.3	-4.6
arc-agi	55.5	+2.6
frontiercode	50.9	-2.1
mmmu-pro	55.0	+2.0
gpqa	54.1	+1.2
frontiermath	52.0	-0.9
aime	53.8	+0.8
simpleqa	53.8	+0.8
livebench-data	53.7	+0.8
livebench-coding	52.2	-0.8
swe-rebench	52.3	-0.7
epoch-game-puzzles	52.4	-0.5
livebench-language	53.4	+0.5
hle	52.4	-0.5
mmlu-pro	53.4	+0.4
livebench-math	52.5	-0.4
livebench-reasoning	52.6	-0.3
swe-atlas-qna	52.6	-0.3
vibe-code	52.6	-0.3
omniscience	53.2	+0.3
livebench-instructions	53.2	+0.3
vals-excel-modeling	52.7	-0.3
vals-legal-research	52.7	-0.3
vals-finance-agent	53.1	+0.2
arc-agi-3	52.7	-0.2
briefcase	52.8	-0.2
swe-atlas-test-writing	52.8	-0.1
tau-bench	52.8	-0.1
analyst-agent	52.8	-0.1
ifbench	53.1	+0.1
vals-code-migration	52.8	-0.1
gdpval	52.8	-0.1
enterprise-ops	53.0	+0.1
automationbench	52.8	-0.1
harvey	52.8	-0.1
vals-tax-agent	52.9	-0.1
gdp-pdf	53.0	+0.1
swe-atlas-refactoring	52.9	-0.0
itbench	52.9	-0.0
apex-agents	53.0	+0.0
terminal-bench-science	52.9	-0.0
swe-bench-pro	52.9	-0.0
multi-swe-bench	52.9	-0.0
enigma-eval	52.9	-0.0
aa-lcr	52.9	-0.0
livecodebench	52.9	-0.0
critpt	52.9	+0.0
gmmlu	52.9	+0.0
math500	52.9	+0.0

What evidence is missing from the fit?

Across all collected settings for this model: 258 observations, 99 contributing, 0 matched without graph weight and 159 excluded. Counts do not establish rank eligibility.

5 · Benchmark family or source protocol has not been reviewed
1 · Original SciCode grading defects are under review; retained as supporting evidence, excluded from aggregate and capability scores. Corrected SciCode-Verified requires a separate protocol review. Sources: https://arxiv.org/abs/2608.04975 and https://artificialanalysis.ai/evaluations/scicode
1 · Incomplete task coverage: 111 of 113 tasks. A common task set is required for ranking.
1 · Fewer than two distinct model identities remain in this evaluation
3 · Evaluator has not been reviewed for independent admission
53 · No reviewed effort for this catalog observation; any separately collected effort measurement is counted separately
58 · Supporting evidence outside the reviewed capability core
11 · Narrow specialist biological, medical or security evaluation; preserved as raw evidence outside general capability scoring.
7 · Kimi K3 Evaluation Details cites Artificial Analysis; retain as published context, not a new Moonshot comparison.
1 · Kimi K3 Evaluation Details cites GLM release blog or Vals AI, per model; retain as published context, not a new Moonshot comparison.
1 · Kimi K3 Evaluation Details cites GLM release blog, Artificial Analysis or OpenAI, per model; retain as published context, not a new Moonshot comparison.
1 · Kimi K3 Evaluation Details cites FrontierSWE leaderboard; retain as published context, not a new Moonshot comparison. Historical provider claim; secondary and superseded for current FrontierSWE comparisons by the independent Proximal FrontierSWE v2 mean@5 board at https://www.frontierswe.com/. Version-unspecified claims are not v2 results. Excluded from weighted Overall.
1 · Kimi K3 Evaluation Details cites official PostTrainBench results; retain as published context, not a new Moonshot comparison.
1 · Kimi K3 Evaluation Details cites Anthropic or OpenAI release reports; retain as published context, not a new Moonshot comparison.
1 · Kimi K3 Evaluation Details cites official Agents Last Exam leaderboard; retain as published context, not a new Moonshot comparison.
1 · Kimi K3 Evaluation Details cites Artificial Analysis / APEX-Agents leaderboard; retain as published context, not a new Moonshot comparison.
3 · Kimi K3 Evaluation Details cites Vals AI; retain as published context, not a new Moonshot comparison.
3 · Composite index or publisher-specific internal evaluation lacks a stable general-capability comparison unit.
1 · No matched opponent in this evaluation unit
2 · The Qwen card cites an external leaderboard or release report for this comparator; it is not a new Qwen evaluation.
1 · The Qwen card cites an external leaderboard or release report for this comparator; it is not a new Qwen evaluation. Historical provider claim; secondary and superseded for current FrontierSWE comparisons by the independent Proximal FrontierSWE v2 mean@5 board at https://www.frontierswe.com/. Version-unspecified claims are not v2 results. Excluded from weighted Overall.
1 · The GLM card attributes this evaluation to Proximal; it is not a new Z.ai comparison. Historical provider claim; secondary and superseded for current FrontierSWE comparisons by the independent Proximal FrontierSWE v2 mean@5 board at https://www.frontierswe.com/. Version-unspecified claims are not v2 results. Excluded from weighted Overall.
1 · The GLM card attributes this evaluation to Artificial Analysis; it is not a new Z.ai comparison.

Inspect every observation and exclusion →

Benchmark results

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

Benchmark / source	Low	Medium	High	XHigh	Max	Non-reasoning
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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	45% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	Not reported
AA-Briefcase EloArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details AA-Briefcase v1.1 is an agentic knowledge work benchmark developed by Artificial Analysis. AA-Briefcase Elo is a combined metric that aggregates rubric pass rate, analytical quality Elo and presentation Elo · Higher is better · Evaluation results measured independently by Artificial Analysis Unit: elo. Source collected 2026-10-18.	Not reported	Not reported	Not reported	Not reported	1,321.43 Elo Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	77.67% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	45.59% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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. CritPt is under review by its authors following external feedback · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	Not reported	20.86% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	43.96% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	Not reported
GDP.pdf: All-pass RateArtificial 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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	22.8% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	Not reported
GDPval-AA v2.1 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 DeepSeek V4.1 Flash (max) at 1600 · Higher is better · Evaluation results measured independently by Artificial Analysis Unit: elo. Source collected 2026-10-18.	Not reported	Not reported	Not reported	Not reported	1,453.47 Elo Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	92.02% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	48.66% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	62.24% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	45.56% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	28.75 Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	48.83% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	Not reported
SciCodeArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Benchmark curated by scientists across 16 disciplines. SciCode is under review following an independent audit of the dataset · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	Not reported	54.4% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	94.44% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	34.23% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	58.33% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	84.64% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	Not reported	Not reported	Not reported	Not reported	21.72% Reported settings Claude Opus 4.8 (Max); Model source: https://artificialanalysis.ai/models/claude-opus-4-8	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 2026-10-18.	62.22% Reported settings Claude Opus 4.8 (Low); Source model ID: anthropic-opus-4-8-low; 120 tasks	71.67% Reported settings Claude Opus 4.8 (Medium); Source model ID: anthropic-opus-4-8-medium; 120 tasks	72.08% Reported settings Claude Opus 4.8 (High); Source model ID: anthropic-opus-4-8-high; 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 2026-10-18.	40.8% Reported settings claude-opus-4-8; Reported effort: low; 113 unique tasks; 451 attempts; 4 runs	48.7% Reported settings claude-opus-4-8; Reported effort: medium; 113 unique tasks; 452 attempts; 4 runs	51.8% Reported settings claude-opus-4-8; Reported effort: high; 113 unique tasks; 452 attempts; 4 runs	54.4% Reported settings claude-opus-4-8; Reported effort: xhigh; 113 unique tasks; 447 attempts; 4 runs	59% Reported settings claude-opus-4-8; Reported effort: max; 111 unique tasks; 429 attempts; 4 runs Incomplete task coverage: 111 of 113 tasks. A common task set is required for ranking. Source label needs review	Not reported
Terminal-Bench 2.1Terminal-Bench · Claude Code Benchmark details Original public board. Configuration and task coverage may differ; inspect source. Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	78.88% Reported settings Opus 4.8; Reported effort: high; 445 trials; recorded 2026-05-28	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 2026-10-18.	50.83% 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 0.0013	55.37% 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	55.63% 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 0.0073	59.45% 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 0.0033	59.6% 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 0.004	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 2026-10-18.	35.14% 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 0.001	40.46% 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 0.005	41.02% 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 0.008	45.49% 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 0.005	46.5% 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 0.006	Not reported

What these comparisons mean

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

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

Download all effort benchmark results · 8,916 results across 259 models · Collected 2026-10-18

Claude Opus 4.8

Compare with

One configuration, seven capabilities

Agentic

Hard reasoning

Coding

Human pref

Knowledge

Multimodal

Long context

Capability score evidence

Benchmark results