UnifyBench

RankingClaude Fable 5

Claude Fable 5

Compare models

Data updated 22 Oct 2026

124 published benchmark measures · 24 benchmark families contribute across 5 task areas. 4 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 Fable 5 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 Fable 5 · Agentic: 78.7 · SupportedClaude Fable 5 · Hard reasoning: 80.0 · SupportedClaude Fable 5 · Coding: 76.0 · SupportedClaude Fable 5 · Multimodal: 80.9 · Supported

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

Agentic11 families · 1 with independent evidence · Supported78.7

11 core families; 18 direct opponents across 7 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: 74.2–80.8; 0/13 scenarios unsupported. Without one publisher: 72.8–81.8; 0/15 unsupported. Smoothing check: 71.2–82.7; 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 · Supported80.0

4 core families; 59 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: 70.0–84.5; 0/6 scenarios unsupported. Without one publisher: 68.6–81.7; 0/16 unsupported. Smoothing check: 73.1–83.1; 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.

Coding4 families · 2 with independent evidence · Supported76.0

4 core families; 35 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: 70.3–78.5; 0/6 scenarios unsupported. Without one publisher: 74.1–79.6; 0/19 unsupported. Smoothing check: 68.4–80.8; 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 prefNo 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; 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.

    Knowledge2 families · 0 with independent evidence · PreliminaryUnknown

    2 core families; 2 direct opponents across 1 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; 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.

    • Gmmlu: 50.0 observed win share
      Anthropic · Source 1
    • Milu: 50.0 observed win share
      Anthropic · Source 1
    Multimodal3 families · 0 with independent evidence · Supported80.9

    3 core families; 6 direct opponents across 3 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: 77.9–84.8; 2/8 scenarios unsupported. Without one publisher: 80.0–86.5; 1/8 unsupported. Smoothing check: 74.5–84.4; 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 contextNo 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; 5/5 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: 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

      24 contributing families across 5 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 Fable
      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-fable-5.svg
      Benchmark scores & sources

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

      BenchmarkBucketScoreHarnessEvidenceSource-recorded date
      MCP-Atlas · source release snapshot; version not specifiedAgentic84.7%
      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: MCP-Atlas · reviewed 2026-10-18

      		Published configuration

      Effort: Max with fallbacks

      contributes to capability      		lab self-reportReviewed 2026-10-18
      MCPMark-Verified · source release snapshot; version not specifiedSupporting evidence87.4%
      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: MCPMark-Verified · reviewed 2026-10-18

      		Published configuration

      Effort: Max with fallbacks

      		lab self-report

      Supporting evidence outside the reviewed capability core

      		Reviewed 2026-10-18
      MILUKnowledge92.9%
      Reported settings & source

      Mean accuracy11languages; adaptive max; five trials; no tools/custom system prompts.

      Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures.

      Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.18.2 · reviewed 2026-10-18

      		Published configuration

      Effort: Not specified

      contributes to capability      		lab self-reportReviewed 2026-10-18
      MLS-Bench-Lite · source release snapshot; version not specifiedSupporting evidence49.9%
      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.

      Cited comparator result; see the benchmark footnote.

      Comparison limit: The Qwen card cites an external leaderboard or release report for this comparator; it is not a new Qwen evaluation. The source states that Fable 5 results may involve fallback execution; an exact configuration is not established.

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

      		Published configuration

      Effort: Not specified

      		lab self-report

      The Qwen card cites an external leaderboard or release report for this comparator; it is not a new Qwen evaluation. The source states that Fable 5 results may involve fallback execution; an exact configuration is not established.

      		Reviewed 2026-10-18
      MLS-Bench-Lite · source release snapshot; version not specifiedSupporting evidence49.9%
      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: MLS-Bench-Lite · reviewed 2026-10-18

      		Published configuration

      Effort: Max with fallbacks

      		lab self-report

      Supporting evidence outside the reviewed capability core

      		Reviewed 2026-10-18
      MMMU-Pro · source release snapshot; version not specifiedMultimodal81.2%
      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. 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: MMMU-Pro · reviewed 2026-10-18

      		Published configuration

      Effort: Max with fallbacks

      contributes to capability      		lab self-reportReviewed 2026-10-18
      MMMU-Pro · source release snapshot; version not specifiedMultimodal86.5%
      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: MMMU-Pro · reviewed 2026-10-18

      		Published configuration

      Effort: Max with fallbacks

      contributes to capability      		lab self-reportReviewed 2026-10-18
      OfficeQA ProAgentic57.9%
      Reported settings & source

      Databricks evaluation reading documents as images; differs from extracted-text harness.

      Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures.

      Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.15.1 · reviewed 2026-10-18

      		Published configuration

      Effort: Not specified

      		lab self-report

      No matched opponent in this evaluation unit

      		Reviewed 2026-10-18
      OfficeQA Pro · source release snapshot; version not specifiedAgentic69.9%
      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: OfficeQA Pro · reviewed 2026-10-18

      		Published configuration

      Effort: Max with fallbacks

      contributes to capability      		lab self-reportReviewed 2026-10-18
      OmniDocBench · source release snapshot; version not specifiedSupporting evidence89.8%
      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: OmniDocBench · reviewed 2026-10-18

      		Published configuration

      Effort: Max with fallbacks

      		lab self-report

      Supporting evidence outside the reviewed capability core

      		Reviewed 2026-10-18
      OSWorld · 2.0 August2026 task releaseAgentic72.9%
      Reported settings & source

      partial pass@1;108tasks; five runs;1080p;500steps; max effort; Opus4.8grader; task fixes; Fable safety interventions score zero.

      Same-condition reruns; supersedes earlier OSWorld2results and incompatible with previous task files. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures.

      Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.14.3 · reviewed 2026-10-18

      		Published configuration

      Effort: Max

      contributes to capability      		lab self-reportReviewed 2026-10-18
      OSWorld · 2.0 August2026 task releaseAgentic36.1%
      Reported settings & source

      strict pass@1;108tasks; five runs;1080p;500steps; max effort; Opus4.8grader; task fixes; Fable safety interventions score zero.

      Same-condition reruns; supersedes earlier OSWorld2results and incompatible with previous task files. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures.

      Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.14.3 · reviewed 2026-10-18

      		Published configuration

      Effort: Max

      contributes to capability      		lab self-reportReviewed 2026-10-18
      OSWorld 2.0 · 2.0Agentic66.1%
      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: OSWorld 2.0 · reviewed 2026-10-18

      		Published configuration

      Effort: Max with fallbacks

      contributes to capability      		lab self-reportReviewed 2026-10-18
      OSWorld-VerifiedAgentic85.96%OSWorld-Verified reported

      Effort: Not specified

      contributes to capability      		official board2026-08-01
      OSWorld-Verified · source release snapshot; version not specifiedAgentic85%
      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: OSWorld-Verified · reviewed 2026-10-18

      		Published configuration

      Effort: Max with fallbacks

      contributes to capability      		lab self-reportReviewed 2026-10-18
      PaperBench · source release snapshot; version not specifiedSupporting evidence88.8%
      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. BasicAgent Code-Dev; Opus4.6 judge,3runs,12h each.

      First-party reported result.

      Comparison limit: The source states that Fable 5 results may involve fallback execution; an exact configuration is not established.

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

      		Published configuration

      Effort: Not specified

      		lab self-report

      The source states that Fable 5 results may involve fallback execution; an exact configuration is not established.

      		Reviewed 2026-10-18
      PerceptionBench · source release snapshot; version not specifiedSupporting evidence57.2%
      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: PerceptionBench · reviewed 2026-10-18

      		Published configuration

      Effort: Max with fallbacks

      		lab self-report

      Supporting evidence outside the reviewed capability core

      		Reviewed 2026-10-18
      PLawBench · source release snapshot; version not specifiedSupporting evidence70.2%
      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.

      Comparison limit: The source states that Fable 5 results may involve fallback execution; an exact configuration is not established.

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

      		Published configuration

      Effort: Not specified

      		lab self-report

      The source states that Fable 5 results may involve fallback execution; an exact configuration is not established.

      		Reviewed 2026-10-18
      PostTrainBench · source release snapshot; version not specifiedSupporting evidence41.4%
      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: PostTrainBench · reviewed 2026-10-18

      		Published configuration

      Effort: Max with fallbacks

      		lab self-report

      Supporting evidence outside the reviewed capability core

      		Reviewed 2026-10-18
      PostTrainBench · source release snapshot; version not specifiedSupporting evidence41.8%
      Reported settings & source

      GLM-5.3 release table; benchmark-specific harness and reasoning configurations in Evaluation Details.

      First-party reported result.

      Comparison limit: The source column includes fallback execution without a documented exact effort.

      zai-org/GLM-5.3 · Performance table: PostTrainBench · reviewed 2026-10-18

      		Published configuration

      Effort: Not specified

      		lab self-report

      The source column includes fallback execution without a documented exact effort.

      		Reviewed 2026-10-18
      PRBench-Finance · source release snapshot; version not specifiedSupporting evidence55.8%
      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.

      Comparison limit: The source states that Fable 5 results may involve fallback execution; an exact configuration is not established.

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

      		Published configuration

      Effort: Not specified

      		lab self-report

      The source states that Fable 5 results may involve fallback execution; an exact configuration is not established.

      		Reviewed 2026-10-18
      PRBench-Legal · source release snapshot; version not specifiedSupporting evidence57.6%
      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.

      Comparison limit: The source states that Fable 5 results may involve fallback execution; an exact configuration is not established.

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

      		Published configuration

      Effort: Not specified

      		lab self-report

      The source states that Fable 5 results may involve fallback execution; an exact configuration is not established.

      		Reviewed 2026-10-18
      ProgramBench · 166 golden-task subsetSupporting evidence86.3%
      Reported settings & source

      mini-swe-agent without six-hour timeout; excludes34flaky-reference tasks; tests restricted to reference-passing tests; up to1Mcontext.

      Hidden-test pass rate, not fraction of completely solved programs. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures.

      Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.11.1 · reviewed 2026-10-18

      		Published configuration

      Effort: Not specified

      		lab self-report

      Supporting evidence outside the reviewed capability core

      		Reviewed 2026-10-18
      ProgramBench · source release snapshot; version not specifiedSupporting evidence76.8%
      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 GLM release blog or Vals AI, per model; retain as published context, not a new Moonshot comparison.

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

      		Published configuration

      Effort: Max with fallbacks

      		lab self-report

      Kimi K3 Evaluation Details cites GLM release blog or Vals AI, per model; retain as published context, not a new Moonshot comparison.

      		Reviewed 2026-10-18
      ProgramBench (Almost Solved) · source release snapshot; version not specifiedSupporting evidence33%
      Reported settings & source

      GLM-5.3 release table; benchmark-specific harness and reasoning configurations in Evaluation Details.

      First-party reported result.

      Comparison limit: The source column includes fallback execution without a documented exact effort.

      zai-org/GLM-5.3 · Performance table: ProgramBench (Almost Solved) · reviewed 2026-10-18

      		Published configuration

      Effort: Not specified

      		lab self-report

      The source column includes fallback execution without a documented exact effort.

      		Reviewed 2026-10-18