Home Benchmark directory CharXiv Reasoning · Validation split · Benchmark definition and scores CharXiv Reasoning · Validation split · Benchmark definition and scores CharXiv Reasoning · Validation split has benchmark disclosures or scores across 8 model releases.
Benchmark versions and metrics Aliases CharXiv Reasoning
Accuracy · % · Higher is better Latest models by series · reported scores Regular display scope. Rankings are grouped by metric and evaluation protocol. Recorded tool access, budgets, and frameworks are shown separately. Scores remain available for reference when conditions are unspecified.
Accuracy % Conditions not specified CharXiv Reasoning · Validation split · Accuracy · Conditions not specified Model and configuration Score (%) Records and evidence GLM-4.6V Not specified 63.2% GLM-4.6V-Flash Not specified 59.6%
History · 8 releases Claude Sonnet 5 2026-06-30 2 score records for this release, including 2 exact values (may include historical records) Claude Opus 4.8 2026-05-28 2 score records for this release, including 2 exact values (may include historical records) Claude Opus 4.7 2026-04-16 2 score records for this release, including 2 exact values (may include historical records) Claude Sonnet 4.6 2026-02-17 3 score records for this release, including 3 exact values (may include historical records) Claude Opus 4.6 2026-02-05 2 score records for this release, including 2 exact values (may include historical records) GLM-4.6V 2025-12-08 1 score record for this release, including 1 exact value (may include historical records) GLM-4.6V-Flash 2025-12-08 1 score record for this release, including 1 exact value (may include historical records) Gemini 3 Pro 2025-11-18 1 score record for this release, including 1 exact value (may include historical records)