Skip to content

Share prompts Sign in Create team library

The Chat Judge verdicts

Which AI is best at what?

Public benchmark results for 19 tests across 9 fields, charted side by side. Every score links to the leaderboard it came from.

Showing Z.ai's best-ranked model on each benchmark.

1 benchmarkUpdated 19 Oct 2026Overall workOne number for agentic work across finance, coding, law and tax.

Vals Index GLM 5.3 #9 · 59.73%

See charts 3 benchmarksUpdated 19 Oct 2026LegalDrafting, extraction, legal reasoning and US legal research.

Legal Benchmarks leaderboard GLM-5.3 #12 · 69.8%
LegalBench GLM Legal #15 · 86.23%
Legal Research Bench GLM 5.3 #9 · 49.04%

See charts 3 benchmarksUpdated 21 Oct 2026Health & medicalPhysician-graded health conversations, medical coding and exam questions.

HealthBench (independent run) GLM-5.3 (max) #11 · 54.5%
MedQA GLM 5 #18 · 94.27%

See charts 2 benchmarksUpdated 19 Oct 2026FinanceAnalyst research on SEC filings and long credit agreements.

Finance Agent GLM 5.3 #9 · 65.00%
CorpFin v2 GLM 5.2 #22 · 66.12%

See charts 2 benchmarksUpdated 1 Sep 2026CodingFixing real GitHub issues and solving fresh programming problems.

SWE-bench Verified GLM 5.3 #6 · 95.4%

See charts 2 benchmarksUpdated 19 Oct 2026Agents & tool useMulti-step tasks an agent completes alone in a terminal.

Terminal-Bench 4.0 GLM 5.3 #12 · 38.89%

See charts 4 benchmarksUpdated 21 Oct 2026Reasoning & knowledgeExpert exams, graduate science, contest maths and broad knowledge.

AIME GLM 4.7 #16 · 93.33%

See charts 1 benchmarkUpdated 20 Oct 2026Chat preferenceWhich answers real people prefer in blind side-by-side votes.

LMArena Text Arena GLM-5.3 (Max) #26 · 1481

ChatJudge does not run these tests. Scores are copied as published by each source, with a link and the date we read them, and only change when we refresh from the source. Different leaderboards use different settings, so compare models within one chart, not across charts. Spotted a mistake? Tell us.