RankingEffort benchmarks
Claude Opus 4
9 measured results · 4 effort settings · 6 benchmark/harness combinations
Change model
Capability profile
One configuration, seven capabilities
- AClaude Opus 4 · Thinking1 of 7 capabilities supported
Gaps are unknown, not zero. Hollow points are preliminary.
Agentic
UnknownHard reasoning
A7.2SupportedCoding
A0.6PreliminaryHuman pref
UnknownKnowledge
UnknownMultimodal
UnknownLong context
Unknown
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 · Thinking: 1.2 overall comparison score. Meets the overall evidence requirements.
Which sources carry the weight?
- LiveCodeBench · 39.5% direct comparison weight · independent of the model provider
- Scale · 30.9% direct comparison weight · independent of the model provider
- Epoch AI · 29.6% direct comparison weight · independent of the model provider
Which benchmark families carry the weight?
- LiveCodeBench · 39.5% direct comparison weight
- Humanity’s Last Exam · 30.9% direct comparison weight
- GPQA · 29.6% 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 2026-11-22.
Both ranks use the same surviving configurations for each removal. The cohort starts with 602 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.
| Removed source | Rank before → after | Rank movement | Cohort remaining / lost support | Score after (change) |
|---|---|---|---|---|
| Artificial Analysis | 219 → 215 | Up 4 | 230 remaining / 372 lost | 1.2 |
| ARC Prize | 435 → 555 | Down 120 | 602 remaining / 0 lost | 0.3 |
| Datacurve | 435 → 435 | No movement | 602 remaining / 0 lost | 1.0 |
| LiveCodeBench | 435 → 407 | Up 28 | 602 remaining / 0 lost | 1.8 |
| Scale | 435 → 443 | Down 8 | 602 remaining / 0 lost | 1.3 |
| Terminal-Bench | 435 → 435 | No movement | 602 remaining / 0 lost | 1.2 |
| Cognition | 435 → 433 | Up 2 | 602 remaining / 0 lost | 1.2 |
| LiveBench | 434 → 434 | No movement | 601 remaining / 1 lost | 1.3 |
| SWE-rebench | 435 → 437 | Down 2 | 602 remaining / 0 lost | 1.2 |
| Epoch AI | 421 → 482 | Down 61 | 588 remaining / 14 lost | 0.5 |
| Vals AI | 434 → 436 | Down 2 | 601 remaining / 1 lost | 1.2 |
| Provider reports · Anthropic | 435 → 435 | No movement | 602 remaining / 0 lost | 1.2 |
| Provider reports · Moonshot AI | 435 → 434 | Up 1 | 602 remaining / 0 lost | 1.2 |
| Provider reports · Meta | 435 → 434 | Up 1 | 602 remaining / 0 lost | 1.2 |
| Provider reports · Mistral | 435 → 435 | No movement | 602 remaining / 0 lost | 1.2 |
| Provider reports · OpenAI | 435 → 435 | No movement | 602 remaining / 0 lost | 1.3 |
| All provider reports | 435 → 434 | Up 1 | 602 remaining / 0 lost | 1.2 |
Lost support means a configuration no longer has a comparison path to the complete reference panel. Its rank is unavailable. A surviving path alone does not establish leaderboard eligibility.
Overlapping sources may remain correlated even with source balancing. Leave-one-source-out movement is a sensitivity diagnostic, not a spread of source-specific ranks or a confidence measure. Scoring methodology · Dated configuration audit
Every benchmark family
| Removed evidence | Recomputed score | Change |
|---|---|---|
| arc-agi | 0.3 | -1.0 |
| gpqa | 0.5 | -0.7 |
| livecodebench | 1.8 | +0.5 |
| simpleqa | 1.6 | +0.3 |
| deepswe | 1.0 | -0.3 |
| aime | 1.4 | +0.2 |
| epoch-game-puzzles | 1.1 | -0.1 |
| mmmu-pro | 1.3 | +0.1 |
| arc-agi-3 | 1.2 | -0.1 |
| mmlu-pro | 1.3 | +0.0 |
| frontiercode | 1.2 | -0.0 |
| livebench-instructions | 1.3 | +0.0 |
| hle | 1.2 | -0.0 |
| multi-swe-bench | 1.2 | -0.0 |
| swe-bench-pro | 1.3 | +0.0 |
| frontiermath | 1.3 | +0.0 |
| livebench-language | 1.3 | +0.0 |
| aa-lcr | 1.2 | -0.0 |
| livebench-coding | 1.2 | -0.0 |
| swe-atlas-refactoring | 1.3 | +0.0 |
| gmmlu | 1.2 | -0.0 |
| enigma-eval | 1.3 | +0.0 |
| vibe-code | 1.2 | -0.0 |
| swe-atlas-qna | 1.2 | -0.0 |
| terminal-bench | 1.2 | -0.0 |
| apex-agents | 1.3 | +0.0 |
| vals-code-migration | 1.2 | -0.0 |
| automationbench | 1.2 | -0.0 |
| livebench-reasoning | 1.2 | -0.0 |
| briefcase | 1.2 | -0.0 |
| vals-finance-agent | 1.2 | -0.0 |
| gdp-pdf | 1.3 | +0.0 |
| vals-legal-research | 1.2 | -0.0 |
| vals-excel-modeling | 1.2 | -0.0 |
| tau-bench | 1.3 | +0.0 |
| vals-tax-agent | 1.2 | -0.0 |
| itbench | 1.2 | -0.0 |
| swe-atlas-test-writing | 1.2 | -0.0 |
| gdpval | 1.3 | +0.0 |
| analyst-agent | 1.3 | +0.0 |
| ifbench | 1.2 | -0.0 |
| omniscience | 1.2 | -0.0 |
| terminal-bench-science | 1.3 | +0.0 |
| enterprise-ops | 1.2 | -0.0 |
| swe-rebench | 1.2 | -0.0 |
| math500 | 1.2 | -0.0 |
| livebench-data | 1.2 | +0.0 |
| livebench-math | 1.2 | +0.0 |
| harvey | 1.2 | +0.0 |
| critpt | 1.2 | +0.0 |
What evidence is missing from the fit?
Across all collected settings for this model: 9 observations, 7 contributing, 0 matched without graph weight and 2 excluded. Counts do not establish rank eligibility.
- 2 · Conflicting values for this configuration in the same evaluation
Benchmark results
Reviewed retained observations are in use. October 2026 Real-SWE board has no 95% confidence whisker, so these scores stay at the 2026-10-18 retrieval. Original source retrieval 2026-10-18; source updated 2026-10-18.
| Benchmark / source | Thinking | Thinking 1K | Thinking 8K | Thinking 16K |
|---|---|---|---|---|
ARC-AGI-2Benchmark detailsOriginal public board. Configuration and task coverage may differ; inspect source. Unit: percent. Source collected 2026-11-22. | Not reported | 0% Reported settingsClaude Opus 4 (Thinking 1K); Source model ID: claude-opus-4-20250514-thinking-1k; 120 tasks | 4.52% Reported settingsClaude Opus 4 (Thinking 8K); Source model ID: claude-opus-4-20250514-thinking-8k; 120 tasks | 8.61% Reported settingsClaude Opus 4 (Thinking 16K); Source model ID: claude-opus-4-20250514-thinking-16k; 120 tasks |
LiveCodeBenchBenchmark detailsEvaluation window 2024-08-01 to 2025-05-01; 454 shared tasks. Unit: percent. Source collected 2026-11-22. | 56.6% Reported settingsClaude-Opus-4 (Thinking); Exact source model: claude-opus-4-20250514; 454 tasks | Not reported | Not reported | Not reported |
HLE · text only · Scale official evaluationBenchmark detailsScale text-only public HLE subset; shares the full HLE family budget; automatic answer judging. Unit: percent. Source collected 2026-11-22. | 10.8% Reported settingsScale text-only public HLE subset; shares the full HLE family budget; automatic answer judging.; published row 2025-05-24T06:34:16.000Z; reported margin ±1.31 percentage points; Potential contamination warning: This model was evaluated after the public release of HLE, allowing model builder access to the prompts and solutions. Sampled at temperature 0.3. | Not reported | Not reported | Not reported |
HLE · full · Scale official evaluationBenchmark detailsScale full public HLE; 2,500 questions; automatic answer judging; temperature 0 when configurable. Public-set contamination caveats retained per result. Unit: percent. Source collected 2026-11-22. | 10.72% Reported settingsScale full public HLE; 2,500 questions; automatic answer judging; temperature 0 when configurable. Public-set contamination caveats retained per result.; published row 2025-05-24T06:35:10.000Z; reported margin ±1.21 percentage points; Potential contamination warning: This model was evaluated after the public release of HLE, allowing model builder access to the prompts and solutions. | Not reported | Not reported | Not reported |
Epoch · GPQA diamond · v1.0.0Benchmark detailsEpoch 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 2026-11-22. | 76.26% Reported settingsEpoch AI Inspect; task 1.0.0; scorer choice; model=claude-opus-4-20250514_16K; run=CtxFusPxCraL3EeRJu2c5k; mean accuracy; https://epoch.ai/data/benchmarks.csv | Not reported | Not reported | Not reported |
Epoch · OTIS Mock AIME 2024-2025 · v1.0.0Benchmark detailsEpoch 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 2026-11-22. | 64.44% Reported settingsEpoch AI Inspect; task 1.0.0; scorer model_graded; model=claude-opus-4-20250514_27K; run=VrRccga4uRVuCjNp637MeY; mean accuracy; https://epoch.ai/data/benchmarks.csv 60% Reported settingsEpoch AI Inspect; task 1.0.0; scorer model_graded; model=claude-opus-4-20250514_16K; run=Xz2F28L5k8HFNfY7Lo8Ggw; mean accuracy; https://epoch.ai/data/benchmarks.csv | 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 · 10,184 results across 381 models · Collected 2026-11-22