RankingEffort benchmarks

GPT-6.1 Terra

Compare models

72 measured results · 6 effort settings · 17 benchmark/harness combinations

Change model

Compare with

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

Compare with

Capability profile

One configuration, seven capabilities

Comparison score · 0–100
GPT-6.1 Terra · Max capability radarComparison scores from 0 to 100. Unknown capabilities have no point; lines stop at gaps. Hollow points are preliminary. Exact values and support labels follow the chart.AgenticHard reasoningCodingHuman prefKnowledgeMultimodalLong context50100GPT-6.1 Terra · Max · Agentic: 72.1 · SupportedGPT-6.1 Terra · Max · Hard reasoning: 74.9 · SupportedGPT-6.1 Terra · Max · Coding: 86.6 · PreliminaryGPT-6.1 Terra · Max · Knowledge: 64.9 · SupportedGPT-6.1 Terra · Max · Multimodal: 62.9 · PreliminaryGPT-6.1 Terra · Max · Long context: 79.2 · Preliminary

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

  • Agentic

    A72.1Supported
  • Hard reasoning

    A74.9Supported
  • Coding

    A86.6Preliminary
  • Human pref

    Unknown
  • Knowledge

    A64.9Supported
  • Multimodal

    A62.9Preliminary
  • Long context

    A79.2Preliminary

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

GPT-6.1 Terra · Max: 77.4 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?
  • Epoch AI · 46.2% direct comparison weight · independent of the model provider
  • Artificial Analysis · 32.2% direct comparison weight · independent of the model provider
  • Terminal-Bench · 21.6% direct comparison weight · independent of the model provider
Which benchmark families carry the weight?
  • Terminal-Bench · 23.1% direct comparison weight
  • GPQA · 13.3% direct comparison weight
  • Aime · 13.1% direct comparison weight
  • simpleqa · 10.9% direct comparison weight
  • FrontierMath · 8.9% direct comparison weight
  • Other families (10) · 30.7% 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.

What changes when a source is removed?
Removed sourceRank before → afterRank movementCohort remaining / lost supportScore after (change)
Artificial Analysis41 → 37Up 4230 remaining / 372 lost79.2 (+1.8 points)
ARC Prize45 → 46Down 1602 remaining / 0 lost79.8 (+2.4 points)
Datacurve45 → 44Up 1602 remaining / 0 lost75.2 (-2.2 points)
LiveCodeBench45 → 44Up 1602 remaining / 0 lost77.4 (-0.0 points)
Scale45 → 47Down 2602 remaining / 0 lost77.9 (+0.6 points)
Terminal-Bench45 → 47Down 2602 remaining / 0 lost75.2 (-2.2 points)
Cognition45 → 41Up 4602 remaining / 0 lost76.1 (-1.3 points)
LiveBench45 → 43Up 2601 remaining / 1 lost78.7 (+1.3 points)
SWE-rebench45 → 45No movement602 remaining / 0 lost77.1 (-0.3 points)
Epoch AI45 → 44Up 1588 remaining / 14 lost78.0 (+0.6 points)
Vals AI45 → 39Up 6601 remaining / 1 lost78.0 (+0.6 points)
Provider reports · Anthropic45 → 46Down 1602 remaining / 0 lost77.3 (-0.1 points)
Provider reports · Moonshot AI45 → 44Up 1602 remaining / 0 lost77.1 (-0.3 points)
Provider reports · Meta45 → 44Up 1602 remaining / 0 lost77.4 (-0.0 points)
Provider reports · Mistral45 → 45No movement602 remaining / 0 lost77.4 (+0.0 points)
Provider reports · OpenAI45 → 43Up 2602 remaining / 0 lost77.4 (+0.0 points)
All provider reports45 → 44Up 1602 remaining / 0 lost77.0 (-0.4 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 evidenceRecomputed scoreChange
simpleqa80.7+3.3
terminal-bench74.7-2.6
deepswe75.2-2.2
arc-agi79.2+1.8
frontiermath75.9-1.5
frontiercode76.1-1.3
aime76.3-1.1
mmmu-pro77.9+0.6
omniscience77.9+0.5
terminal-bench-science76.9-0.5
epoch-game-puzzles76.9-0.5
briefcase77.8+0.4
gpqa77.0-0.4
enigma-eval77.7+0.3
hle77.7+0.3
livebench-instructions77.7+0.3
tau-bench77.1-0.3
ifbench77.1-0.3
gdpval77.7+0.3
vibe-code77.1-0.3
swe-rebench77.1-0.3
vals-code-migration77.2-0.2
vals-finance-agent77.6+0.2
mmlu-pro77.6+0.2
swe-atlas-qna77.2-0.2
vals-excel-modeling77.2-0.2
swe-atlas-test-writing77.2-0.2
gdp-pdf77.2-0.2
critpt77.6+0.2
livebench-coding77.2-0.2
vals-tax-agent77.5+0.1
vals-legal-research77.5+0.1
enterprise-ops77.5+0.1
arc-agi-377.5+0.1
livebench-language77.5+0.1
livebench-reasoning77.3-0.1
multi-swe-bench77.3-0.1
livebench-data77.3-0.1
swe-bench-pro77.3-0.1
swe-atlas-refactoring77.3-0.0
livecodebench77.3-0.0
harvey77.4-0.0
apex-agents77.4+0.0
itbench77.4-0.0
automationbench77.4+0.0
aa-lcr77.4+0.0
analyst-agent77.4+0.0
math50077.4-0.0
gmmlu77.4-0.0
livebench-math77.4-0.0
What evidence is missing from the fit?

Across all collected settings for this model: 72 observations, 72 contributing, 0 matched without graph weight and 0 excluded. Counts do not establish rank eligibility.

    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.

    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 / sourceLowMediumHighXHighMaxNon-reasoning
    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 2026-11-22.

    1,031.95 Elo
    Reported settings

    GPT-6.1 Terra (Low); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-low

    1,190.66 Elo
    Reported settings

    GPT-6.1 Terra (Medium); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-medium

    1,265.03 Elo
    Reported settings

    GPT-6.1 Terra (High); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-high

    1,333.66 Elo
    Reported settings

    GPT-6.1 Terra (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-xhigh

    1,399.69 Elo
    Reported settings

    GPT-6.1 Terra (Max); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra

    1,225.88 Elo
    Reported settings

    GPT-6.1 Terra (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-non-reasoning

    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 2026-11-22.

    81%
    Reported settings

    GPT-6.1 Terra (Low); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-low

    80.67%
    Reported settings

    GPT-6.1 Terra (Medium); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-medium

    82.33%
    Reported settings

    GPT-6.1 Terra (High); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-high

    81.67%
    Reported settings

    GPT-6.1 Terra (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-xhigh

    82.67%
    Reported settings

    GPT-6.1 Terra (Max); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra

    67.67%
    Reported settings

    GPT-6.1 Terra (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-non-reasoning

    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 2026-11-22.

    56.25%
    Reported settings

    GPT-6.1 Terra (Low); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-low

    57.8%
    Reported settings

    GPT-6.1 Terra (Medium); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-medium

    58.46%
    Reported settings

    GPT-6.1 Terra (High); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-high

    59.93%
    Reported settings

    GPT-6.1 Terra (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-xhigh

    60.17%
    Reported settings

    GPT-6.1 Terra (Max); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra

    51.04%
    Reported settings

    GPT-6.1 Terra (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-non-reasoning

    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 2026-11-22.

    15.14%
    Reported settings

    GPT-6.1 Terra (Low); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-low

    21.43%
    Reported settings

    GPT-6.1 Terra (Medium); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-medium

    25.71%
    Reported settings

    GPT-6.1 Terra (High); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-high

    26.57%
    Reported settings

    GPT-6.1 Terra (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-xhigh

    28%
    Reported settings

    GPT-6.1 Terra (Max); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra

    4.86%
    Reported settings

    GPT-6.1 Terra (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-non-reasoning

    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 2026-11-22.

    25.6%
    Reported settings

    GPT-6.1 Terra (Low); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-low

    28.4%
    Reported settings

    GPT-6.1 Terra (Medium); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-medium

    28%
    Reported settings

    GPT-6.1 Terra (High); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-high

    26.6%
    Reported settings

    GPT-6.1 Terra (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-xhigh

    28.6%
    Reported settings

    GPT-6.1 Terra (Max); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra

    23.2%
    Reported settings

    GPT-6.1 Terra (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-non-reasoning

    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 2026-11-22.

    1,203.72 Elo
    Reported settings

    GPT-6.1 Terra (Low); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-low

    1,324.24 Elo
    Reported settings

    GPT-6.1 Terra (Medium); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-medium

    1,353.41 Elo
    Reported settings

    GPT-6.1 Terra (High); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-high

    1,416.43 Elo
    Reported settings

    GPT-6.1 Terra (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-xhigh

    1,475.84 Elo
    Reported settings

    GPT-6.1 Terra (Max); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra

    1,285.53 Elo
    Reported settings

    GPT-6.1 Terra (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-non-reasoning

    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 2026-11-22.

    35.82%
    Reported settings

    GPT-6.1 Terra (Low); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-low

    40.22%
    Reported settings

    GPT-6.1 Terra (Medium); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-medium

    43.28%
    Reported settings

    GPT-6.1 Terra (High); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-high

    43.74%
    Reported settings

    GPT-6.1 Terra (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-xhigh

    46.25%
    Reported settings

    GPT-6.1 Terra (Max); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra

    21.13%
    Reported settings

    GPT-6.1 Terra (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-non-reasoning

    MLCR-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
    Benchmark details

    Share of responses judged complete with no claim contradicting the record · Expert + compound (hard) sets · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

    Unit: percent. Source collected 2026-11-22.

    Not reportedNot reportedNot reportedNot reported
    30.56%
    Reported settings

    GPT-6.1 Terra (Max); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra

    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 2026-11-22.

    82.49%
    Reported settings

    GPT-6.1 Terra (Low); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-low

    82.77%
    Reported settings

    GPT-6.1 Terra (Medium); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-medium

    83.82%
    Reported settings

    GPT-6.1 Terra (High); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-high

    83.82%
    Reported settings

    GPT-6.1 Terra (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-xhigh

    84.68%
    Reported settings

    GPT-6.1 Terra (Max); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra

    76.01%
    Reported settings

    GPT-6.1 Terra (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-non-reasoning

    AA-Omniscience AccuracyArtificial Analysis · 3000 questions; 1 attempt; correct-answer share; GPT-5.6 Luna medium grader
    Benchmark details

    AA-Omniscience v1.1 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 2026-11-22.

    45.57%
    Reported settings

    GPT-6.1 Terra (Low); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-low

    47.43%
    Reported settings

    GPT-6.1 Terra (Medium); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-medium

    47.27%
    Reported settings

    GPT-6.1 Terra (High); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-high

    48.03%
    Reported settings

    GPT-6.1 Terra (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-xhigh

    47.63%
    Reported settings

    GPT-6.1 Terra (Max); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra

    40.73%
    Reported settings

    GPT-6.1 Terra (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-non-reasoning

    Terminal-Bench-Science 0.1Artificial Analysis · mini-swe-agent; Terminal-Bench-Science 0.1.0; 70 tasks; pass@1 averaged over 3 repeats
    Benchmark details

    Share of tasks passed (pass@1), averaged over three repeats · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

    Unit: percent. Source collected 2026-11-22.

    7.62%
    Reported settings

    GPT-6.1 Terra (Low); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-low

    24.76%
    Reported settings

    GPT-6.1 Terra (Medium); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-medium

    33.33%
    Reported settings

    GPT-6.1 Terra (High); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-high

    41.9%
    Reported settings

    GPT-6.1 Terra (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-xhigh

    57.62%
    Reported settings

    GPT-6.1 Terra (Max); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra

    24.29%
    Reported settings

    GPT-6.1 Terra (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-non-reasoning

    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 2026-11-22.

    15.15%
    Reported settings

    GPT-6.1 Terra (Low); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-low

    32.32%
    Reported settings

    GPT-6.1 Terra (Medium); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-medium

    41.92%
    Reported settings

    GPT-6.1 Terra (High); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-high

    46.46%
    Reported settings

    GPT-6.1 Terra (Xhigh); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-xhigh

    55.56%
    Reported settings

    GPT-6.1 Terra (Max); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra

    26.77%
    Reported settings

    GPT-6.1 Terra (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-6-1-terra-non-reasoning

    Terminal-Bench 4.0 · codexTerminal-Bench · Codex
    Benchmark details

    Terminal-Bench 4.0; Codex. Open each result for its exact effort, task subset, metric and protocol details.

    Unit: percent. Source collected 2026-11-22.

    Not reportedNot reportedNot reportedNot reported
    51.8%
    Reported settings

    Terminal-Bench 4.0.0; 66 tasks; Codex; effort max; reported margin ±2.8 percentage points, confidence level unspecified; release date 2026-10-26; board checked 2026-11-18

    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 2026-11-22.

    Not reportedNot reportedNot reportedNot reported
    90.88%
    Reported settings

    Epoch AI Inspect; task 2.0.0; scorer verification_code; model=gpt-6.1-terra_max; run=KkEoE6hBF8RB4ff43Xv63e; 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 2026-11-22.

    Not reportedNot reportedNot reportedNot reported
    94.32%
    Reported settings

    Epoch AI Inspect; task 1.0.11; scorer choice; model=gpt-6.1-terra_max; run=5Jm2Rvv9sNx7M8nodinFAp; mean accuracy; https://epoch.ai/data/benchmarks.csv

    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 2026-11-22.

    Not reportedNot reportedNot reportedNot reported
    100%
    Reported settings

    Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=gpt-6.1-terra_max; run=YnwukT2gyJHGKzPNPU9hry; mean accuracy; https://epoch.ai/data/benchmarks.csv

    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 2026-11-22.

    Not reportedNot reportedNot reportedNot reported
    57.3%
    Reported settings

    Epoch AI Inspect; task 1.2.0; scorer simpleqa_scorer; model=gpt-6.1-terra_max; run=Bixeio8A6SejZtUzFXFSXK; mean accuracy; https://epoch.ai/data/benchmarks.csv

    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