RankingEffort benchmarks

Agnes 2.5 Pro Theta

Compare models

11 measured results · 1 effort settings · 11 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
Agnes 2.5 Pro Theta · Reasoning 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 context50100Agnes 2.5 Pro Theta · Reasoning · Agentic: 49.3 · SupportedAgnes 2.5 Pro Theta · Reasoning · Hard reasoning: 41.3 · SupportedAgnes 2.5 Pro Theta · Reasoning · Coding: 44.0 · PreliminaryAgnes 2.5 Pro Theta · Reasoning · Knowledge: 18.6 · PreliminaryAgnes 2.5 Pro Theta · Reasoning · Multimodal: 34.5 · PreliminaryAgnes 2.5 Pro Theta · Reasoning · Long context: 71.2 · Preliminary

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

  • Agentic

    A49.3Supported
  • Hard reasoning

    A41.3Supported
  • Coding

    A44.0Preliminary
  • Human pref

    Unknown
  • Knowledge

    A18.6Preliminary
  • Multimodal

    A34.5Preliminary
  • Long context

    A71.2Preliminary

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

Agnes 2.5 Pro Theta · Reasoning: 35.1 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?
  • Artificial Analysis · 100.0% direct comparison weight · independent of the model provider
Which benchmark families carry the weight?
  • Humanity’s Last Exam · 10.8% direct comparison weight
  • Critpt · 10.6% direct comparison weight
  • Mmmu Pro · 10.4% direct comparison weight
  • Aa Lcr · 9.6% direct comparison weight
  • GDPval · 9.1% direct comparison weight
  • Other families (6) · 49.5% 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 AnalysisComparison support lost; no rank230 remaining / 372 lostUnavailable
ARC Prize143 → 138Up 5602 remaining / 0 lost40.6 (+5.5 points)
Datacurve143 → 145Down 2602 remaining / 0 lost30.9 (-4.2 points)
LiveCodeBench143 → 143No movement602 remaining / 0 lost35.1 (-0.1 points)
Scale143 → 143No movement602 remaining / 0 lost36.9 (+1.8 points)
Terminal-Bench143 → 147Down 4602 remaining / 0 lost35.1 (+0.0 points)
Cognition143 → 149Down 6602 remaining / 0 lost33.4 (-1.7 points)
LiveBench143 → 142Up 1601 remaining / 1 lost36.3 (+1.2 points)
SWE-rebench143 → 142Up 1602 remaining / 0 lost35.3 (+0.2 points)
Epoch AI141 → 142Down 1588 remaining / 14 lost35.7 (+0.6 points)
Vals AI143 → 144Down 1601 remaining / 1 lost34.4 (-0.8 points)
Provider reports · Anthropic143 → 143No movement602 remaining / 0 lost35.1 (-0.0 points)
Provider reports · Moonshot AI143 → 143No movement602 remaining / 0 lost34.6 (-0.5 points)
Provider reports · Meta143 → 143No movement602 remaining / 0 lost35.1 (-0.1 points)
Provider reports · Mistral143 → 143No movement602 remaining / 0 lost35.1 (+0.0 points)
Provider reports · OpenAI143 → 143No movement602 remaining / 0 lost35.3 (+0.2 points)
All provider reports143 → 143No movement602 remaining / 0 lost34.7 (-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
omniscience39.9+4.8
arc-agi39.8+4.7
deepswe30.9-4.2
briefcase31.6-3.5
mmmu-pro37.1+2.0
aa-lcr33.3-1.8
frontiercode33.4-1.7
gdpval33.5-1.6
hle36.2+1.1
gdp-pdf36.1+0.9
terminal-bench-science36.1+0.9
simpleqa36.0+0.8
terminal-bench35.9+0.8
epoch-game-puzzles34.4-0.7
ifbench34.6-0.5
arc-agi-334.6-0.5
automationbench34.6-0.5
enigma-eval35.6+0.5
mmlu-pro35.6+0.5
livebench-instructions35.5+0.4
multi-swe-bench34.8-0.4
livebench-language35.4+0.3
apex-agents35.4+0.3
swe-atlas-qna34.9-0.2
gpqa34.9-0.2
enterprise-ops34.9-0.2
aime34.9-0.2
livebench-coding34.9-0.2
swe-rebench35.3+0.2
vibe-code34.9-0.2
analyst-agent35.3+0.1
tau-bench35.0-0.1
vals-code-migration35.0-0.1
livebench-reasoning35.0-0.1
vals-legal-research35.0-0.1
swe-atlas-test-writing35.0-0.1
harvey35.0-0.1
vals-tax-agent35.0-0.1
vals-finance-agent35.0-0.1
swe-bench-pro35.2+0.1
vals-excel-modeling35.1-0.1
gmmlu35.2+0.1
livecodebench35.1-0.0
itbench35.1-0.0
swe-atlas-refactoring35.1-0.0
critpt35.2+0.0
frontiermath35.1-0.0
livebench-data35.2+0.0
livebench-math35.1-0.0
math50035.1-0.0
What evidence is missing from the fit?

Across all collected settings for this model: 11 observations, 11 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 / sourceReasoning
    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.

    1,495.96 Elo
    Reported settings

    Agnes 2.5 Pro Theta; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/agnes-2-5-pro-theta; Mode from evaluator isReasoning metadata; effort budget not specified

    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.

    84.67%
    Reported settings

    Agnes 2.5 Pro Theta; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/agnes-2-5-pro-theta; Mode from evaluator isReasoning metadata; effort budget not specified

    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.

    48.82%
    Reported settings

    Agnes 2.5 Pro Theta; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/agnes-2-5-pro-theta; Mode from evaluator isReasoning metadata; effort budget not specified

    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

    Agnes 2.5 Pro Theta; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/agnes-2-5-pro-theta; Mode from evaluator isReasoning metadata; effort budget not specified

    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.

    14.2%
    Reported settings

    Agnes 2.5 Pro Theta; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/agnes-2-5-pro-theta; Mode from evaluator isReasoning metadata; effort budget not specified

    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.

    1,507 Elo
    Reported settings

    Agnes 2.5 Pro Theta; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/agnes-2-5-pro-theta; Mode from evaluator isReasoning metadata; effort budget not specified

    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.

    37.16%
    Reported settings

    Agnes 2.5 Pro Theta; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/agnes-2-5-pro-theta; Mode from evaluator isReasoning metadata; effort budget not specified

    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.

    78.44%
    Reported settings

    Agnes 2.5 Pro Theta; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/agnes-2-5-pro-theta; Mode from evaluator isReasoning metadata; effort budget not specified

    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.

    21.33%
    Reported settings

    Agnes 2.5 Pro Theta; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/agnes-2-5-pro-theta; Mode from evaluator isReasoning metadata; effort budget not specified

    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.

    3.81%
    Reported settings

    Agnes 2.5 Pro Theta; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/agnes-2-5-pro-theta; Mode from evaluator isReasoning metadata; effort budget not specified

    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.

    10.1%
    Reported settings

    Agnes 2.5 Pro Theta; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/agnes-2-5-pro-theta; Mode from evaluator isReasoning metadata; effort budget not specified

    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