UnifyBench

RankingClaude Opus 4.8

Claude Opus 4.8

Compare models

Data updated 22 Oct 2026

113 published benchmark measures · 23 benchmark families contribute across 6 task areas. 5 capability estimates available in the full profile. See all results ↓ · Compare published benchmarks →

Model evidence summary

This profile combines published settings. It is not a runnable configuration or a leaderboard rank. Compare measured configurations →

Performance profile

Capabilities

Adjusted comparison score · 0–100
Claude Opus 4.8 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 context50100Claude Opus 4.8 · Agentic: 58.5 · SupportedClaude Opus 4.8 · Hard reasoning: 40.1 · SupportedClaude Opus 4.8 · Coding: 61.0 · SupportedClaude Opus 4.8 · Multimodal: 58.4 · SupportedClaude Opus 4.8 · Long context: 63.1 · Supported

Filled points are supported; hollow points are preliminary. Lines stop at unknown capabilities. Exact values and sources follow below.

Agentic10 families · 0 with independent evidence · Supported58.5

10 core families; 15 direct opponents across 8 labs.

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: 54.8–61.0; 0/13 scenarios unsupported. Without one publisher: 57.1–60.6; 0/15 unsupported. Smoothing check: 57.0–58.5; 0/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: family and publisher holdouts improved on neutral predictions within evaluated coverage; source sensitivity still requires inspection. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

Hard reasoning4 families · 1 with independent evidence · Supported40.1

4 core families; 64 direct opponents across 10 labs.

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: 33.5–44.3; 0/6 scenarios unsupported. Without one publisher: 38.1–41.5; 0/16 unsupported. Smoothing check: 39.4–42.3; 0/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: family and publisher holdouts improved on neutral predictions within evaluated coverage. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

Coding3 families · 2 with independent evidence · Supported61.0

3 core families; 36 direct opponents across 9 labs.

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: 47.8–74.6; 0/6 scenarios unsupported. Without one publisher: 54.9–63.8; 0/19 unsupported. Smoothing check: 58.9–61.4; 0/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: family and publisher holdouts improved on neutral predictions within evaluated coverage. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

Human pref1 families · 1 with independent evidence · PreliminaryUnknown

1 core families; 76 direct opponents across 17 labs. No connected comparison to the complete reference panel

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: No supported estimate; 1/1 scenarios unsupported. Without one publisher: No supported estimate; 1/1 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: cross-family and cross-publisher predictions could not be evaluated from the available comparison network. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

  • Human preference (Arena): 86.8 observed win share
    lmarena.ai · Source 1
KnowledgeNo comparable evidenceUnknown

0 core families; 0 direct opponents across 0 labs. No admitted core comparisons

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: No supported estimate; 4/4 scenarios unsupported. Without one publisher: No supported estimate; 5/5 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: almost no publisher-held-out comparisons remained estimable. Cross-source validity is unresolved. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

    Multimodal2 families · 0 with independent evidence · Supported58.4

    2 core families; 6 direct opponents across 5 labs.

    Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

    Refit without one family: 57.9–59.2; 2/8 scenarios unsupported. Without one publisher: 47.1–59.2; 1/8 unsupported. Smoothing check: 56.7–59.0; 0/3 unsupported. These are sensitivity checks, not confidence intervals.

    Audit: limited publisher-held-out coverage. Cross-source validity is unresolved. Validation audit.

    Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

    Long context3 families · 0 with independent evidence · Supported63.1

    3 core families; 6 direct opponents across 2 labs.

    Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

    Refit without one family: 53.9–65.9; 1/5 scenarios unsupported. Without one publisher: 63.1–63.1; 3/5 unsupported. Smoothing check: 62.5–63.2; 0/3 unsupported. These are sensitivity checks, not confidence intervals.

    Audit: family and publisher holdouts underperformed neutral predictions. Treat this as an exploratory estimate. Validation audit.

    Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

    Scores estimate outcomes against a shared reference panel; they are not accuracy percentages. Sparse or disconnected evidence cannot qualify an overall profile. Open a capability to inspect its evidence.

    Reported effort · Mixed settings

    Settings reported in this model's published benchmark results, including results outside the aggregate. Effort names are provider-specific. These are not API defaults or equal compute budgets.

    Mixed settings means multiple settings occur in the evidence. Best across efforts means the source selected its best reported result across settings; it does not mean Max. Unspecified settings stay unknown. This model-summary chart combines reported settings. The leaderboard keeps identified configurations separate and excludes unknown effort.

    Inspect each result and its source ↓ · Download effort evidence
    Score contributions and missing evidence

    23 contributing families across 6 capabilities. Fixed reference panels do not change when the catalog expands.

    Capability is fitted jointly across families. Capability estimates below describe different task areas; their weighted sum is not the Capability score.

    Results without reviewed compatibility or a reference match remain in the raw evidence below. Coverage counts only contributing results.

    Model information & shareable badge
    Lab
    Anthropic
    Catalog status
    active
    Availability
    Public provider catalog; account and region restrictions may apply
    Family
    Claude Opus
    Released
    —
    Context
    —
    License
    proprietary
    Model card
    https://platform.claude.com/docs/en/about-claude/model-deprecations
    Default Capability family coverage
    Documented-evidence family coverage/badge/claude-opus-4-8.svg
    Benchmark scores & sources

    Original results, evaluation harnesses, and evidence behind this model.

    BenchmarkBucketScoreHarnessEvidenceSource-recorded date
    Toolathlon Verified · source release snapshot; version not specifiedAgentic76.2%
    Reported settings & source

    Meta Model API xhigh; Gemini high; Opus max; GPT xhigh. Best of self-reported or Meta reproduction unless specified. OSWorld GUI only; Terminal bash only5attempts89tasks6CPU8GB; DeepSWE no internet5attempts.

    First-party reported result; comparator results retain the source evaluation setup.

    Muse Spark 1.1 evaluation report Figure44 · Figure44 physical page101; methodology pages101–105 · reviewed 2026-09-06

    		Published configuration

    Effort: Not specified

    contributes to capability    		lab self-reportReviewed 2026-09-06
    Toolathlon Verified (Pass@1) · source release snapshot; version not specifiedAgentic76.2%
    Reported settings & source

    Source benchmark-specific methodology; Qwen benchmark effort unspecified, GPT-5.6 Sol max per table header. Max in Qwen model names is not a reported effort setting. Comparator settings and source citations as documented in the model card.

    First-party reported result.

    Qwen/Qwen3.8-2.4T-A95B · Performance table: Toolathlon Verified (Pass@1) · reviewed 2026-10-18

    		Published configuration

    Effort: Not specified

    contributes to capability    		lab self-reportReviewed 2026-10-18
    Toolathlon-Verified · source release snapshot; version not specifiedAgentic76.2%
    Reported settings & source

    Table headers: Kimi K3, GPT-5.6 Sol, Opus 4.8 and GLM-5.2 max; GPT-5.5 xhigh; Fable 5 max with fallbacks. Benchmark-specific footnotes override header settings; cited results are not new Moonshot evaluations. Benchmark-specific tools, harness and budget documented under Evaluation Details. Kimi temp1; single-step top_p.95,agentic top_p1; vision tools=Python,3runs except ZeroBench5runs.

    First-party reported result.

    moonshotai/Kimi-K3 · Performance table: Toolathlon-Verified · reviewed 2026-10-18

    		Published configuration

    Effort: Max

    contributes to capability    		lab self-reportReviewed 2026-10-18
    Toolathlon-Verified · source release snapshot; version not specifiedAgentic76.2%
    Reported settings & source

    DeepSeek updated 0813/0731 release evaluation table; tool/harness settings per model card.

    First-party reported result.

    deepseek-ai/DeepSeek-V4-Pro-0813 · Performance table: Toolathlon-Verified · reviewed 2026-10-18

    		Published configuration

    Effort: Not specified

    contributes to capability    		lab self-reportReviewed 2026-10-18
    Toolathlon-Verified (official board)Supporting evidence76.2%Toolathlon-Verified Default agent

    Effort: Not specified

    		official board

    Supporting evidence outside the reviewed capability core

    		2026-06-30
    Video-MME (w. sub) · source release snapshot; version not specifiedSupporting evidence86%
    Reported settings & source

    Table headers: Kimi K3, GPT-5.6 Sol, Opus 4.8 and GLM-5.2 max; GPT-5.5 xhigh; Fable 5 max with fallbacks. Benchmark-specific footnotes override header settings; cited results are not new Moonshot evaluations. Benchmark-specific tools, harness and budget documented under Evaluation Details. Kimi temp1; single-step top_p.95,agentic top_p1; vision tools=Python,3runs except ZeroBench5runs.

    First-party reported result.

    moonshotai/Kimi-K3 · Performance table: Video-MME (w. sub) · reviewed 2026-10-18

    		Published configuration

    Effort: Max

    		lab self-report

    Supporting evidence outside the reviewed capability core

    		Reviewed 2026-10-18
    WebArena Verified · source release snapshot; version not specifiedAgentic71.2%
    Reported settings & source

    Meta Model API xhigh; Gemini high; Opus max; GPT xhigh. Best of self-reported or Meta reproduction unless specified. OSWorld GUI only; Terminal bash only5attempts89tasks6CPU8GB; DeepSWE no internet5attempts.

    First-party reported result; comparator results retain the source evaluation setup.

    Muse Spark 1.1 evaluation report Figure44 · Figure44 physical page101; methodology pages101–105 · reviewed 2026-09-06

    		Published configuration

    Effort: Not specified

    contributes to capability    		lab self-reportReviewed 2026-09-06
    WideSearch · source release snapshot; version not specifiedSupporting evidence72.9%
    Reported settings & source

    Source benchmark-specific methodology; Qwen benchmark effort unspecified, GPT-5.6 Sol max per table header. Max in Qwen model names is not a reported effort setting. Comparator settings and source citations as documented in the model card. Item-F1 over4runs; Qwen-Agent for Qwen,ClaudeCode for comparators.

    First-party reported result.

    Qwen/Qwen3.8-2.4T-A95B · Performance table: WideSearch · reviewed 2026-10-18

    		Published configuration

    Effort: Not specified

    		lab self-report

    Supporting evidence outside the reviewed capability core

    		Reviewed 2026-10-18
    WorkSpaceBench · source release snapshot; version not specifiedSupporting evidence66.8%
    Reported settings & source

    Source benchmark-specific methodology; Qwen benchmark effort unspecified, GPT-5.6 Sol max per table header. Max in Qwen model names is not a reported effort setting. Comparator settings and source citations as documented in the model card.

    First-party reported result.

    Qwen/Qwen3.8-2.4T-A95B · Performance table: WorkSpaceBench · reviewed 2026-10-18

    		Published configuration

    Effort: Not specified

    		lab self-report

    Supporting evidence outside the reviewed capability core

    		Reviewed 2026-10-18
    WorldVQA ForceAnswer · source release snapshot; version not specifiedSupporting evidence39.1%
    Reported settings & source

    Table headers: Kimi K3, GPT-5.6 Sol, Opus 4.8 and GLM-5.2 max; GPT-5.5 xhigh; Fable 5 max with fallbacks. Benchmark-specific footnotes override header settings; cited results are not new Moonshot evaluations. Benchmark-specific tools, harness and budget documented under Evaluation Details. Kimi temp1; single-step top_p.95,agentic top_p1; vision tools=Python,3runs except ZeroBench5runs.

    First-party reported result.

    moonshotai/Kimi-K3 · Performance table: WorldVQA ForceAnswer · reviewed 2026-10-18

    		Published configuration

    Effort: Max

    		lab self-report

    Supporting evidence outside the reviewed capability core

    		Reviewed 2026-10-18
    ZeroBench (pass@5) · source release snapshot; version not specifiedSupporting evidence17%
    Reported settings & source

    Table headers: Kimi K3, GPT-5.6 Sol, Opus 4.8 and GLM-5.2 max; GPT-5.5 xhigh; Fable 5 max with fallbacks. Benchmark-specific footnotes override header settings; cited results are not new Moonshot evaluations. Benchmark-specific tools, harness and budget documented under Evaluation Details. without tools Kimi temp1; single-step top_p.95,agentic top_p1; vision tools=Python,3runs except ZeroBench5runs.

    First-party reported result.

    moonshotai/Kimi-K3 · Performance table: ZeroBench (pass@5) · reviewed 2026-10-18

    		Published configuration

    Effort: Max

    		lab self-report

    Supporting evidence outside the reviewed capability core

    		Reviewed 2026-10-18
    ZeroBench (pass@5) · source release snapshot; version not specifiedSupporting evidence34%
    Reported settings & source

    Table headers: Kimi K3, GPT-5.6 Sol, Opus 4.8 and GLM-5.2 max; GPT-5.5 xhigh; Fable 5 max with fallbacks. Benchmark-specific footnotes override header settings; cited results are not new Moonshot evaluations. Benchmark-specific tools, harness and budget documented under Evaluation Details. with tools Kimi temp1; single-step top_p.95,agentic top_p1; vision tools=Python,3runs except ZeroBench5runs.

    First-party reported result.

    moonshotai/Kimi-K3 · Performance table: ZeroBench (pass@5) · reviewed 2026-10-18

    		Published configuration

    Effort: Max

    		lab self-report

    Supporting evidence outside the reviewed capability core

    		Reviewed 2026-10-18
    τ³-Banking · source release snapshot; version not specifiedSupporting evidence27.6%
    Reported settings & source

    Table headers: Kimi K3, GPT-5.6 Sol, Opus 4.8 and GLM-5.2 max; GPT-5.5 xhigh; Fable 5 max with fallbacks. Benchmark-specific footnotes override header settings; cited results are not new Moonshot evaluations. Benchmark-specific tools, harness and budget documented under Evaluation Details. Kimi temp1; single-step top_p.95,agentic top_p1; vision tools=Python,3runs except ZeroBench5runs.

    Cited result; see benchmark-specific Evaluation Details.

    Comparison limit: Kimi K3 Evaluation Details cites Artificial Analysis; retain as published context, not a new Moonshot comparison.

    moonshotai/Kimi-K3 · Performance table: τ³-Banking · reviewed 2026-10-18

    		Published configuration

    Effort: Max

    		lab self-report

    Kimi K3 Evaluation Details cites Artificial Analysis; retain as published context, not a new Moonshot comparison.

    		Reviewed 2026-10-18
    GDPval-AAAgentic————
    GPQA DiamondHard reasoning————
    Humanity's Last ExamHard reasoning————
    LiveCodeBenchCoding————
    MMLU-ProKnowledge————
    OSWorld-VerifiedAgentic————
    SWE-bench ProAgentic————
    SWE-bench VerifiedAgentic————
    Terminal-Bench 2.1Agentic————