The Model Index
Up to four models

Claude Opus 4.8 vs GLM-5.3

On the index (GPT-4 = 100), Claude Opus 4.8 scores 170 and GLM-5.3 168. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to Claude Opus 4.8
Presets
Claude Opus 4.8
170154–186
WeightsClosed
ReleasedMay 2026
Price in / out$5.0 / $25
Context1M
Core tests taken6 of 9
GLM-5.3
168152–184
WeightsOpen
ReleasedAug 2026
Price in / out$0.070 / $1.4
Context1.05M
Core tests taken5 of 9

Claude Opus 4.8 and GLM-5.3 are too close to call (2 points apart, within the uncertainty), ahead on 2 of 5 shared core tests, at 25× the price.

Test by testthird-party results

Claude Opus 4.8GLM-5.3
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)LiveBench Data AnalysisLiveBench Data AnalysisLiveBench IFLiveBench IFLiveBench LanguageLiveBench LanguageLiveBench ReasoningLiveBench ReasoningGPQA Diamond (Vals AI)GPQA Diamond (Vals AI)MMLU ProMMLU ProSimpleBenchSimpleBenchSimpleQA VerifiedSimpleQA VerifiedMATHSFrontierMath (tiers 1-3)FrontierMath (tiers 1-3)AIME-style maths (OTIS mock)AIME-style maths (OTIS mock)LiveBench MathematicsLiveBench MathematicsChess puzzlesChess puzzlesFrontierMath (tier 4)FrontierMath (tier 4)CODINGTerminal-Bench 4.0Terminal-Bench 4.0LiveBench Agentic CodingLiveBench Agentic CodingLiveBench CodingLiveBench CodingCode MigrationCode MigrationLiveCodeBenchLiveCodeBenchProgramBenchProgramBenchSWE-benchSWE-benchVibe Code Bench v1.1Vibe Code Bench v1.1DeepSWEDeepSWEWeirdMLWeirdMLAGENTSAPEX-AgentsAPEX-AgentsSkillsBenchSkillsBenchTerminal-Bench-ScienceTerminal-Bench-ScienceDTBenchDTBenchLMCALMCAMystery game puzzlesMystery game puzzlesPROFESSIONAL WORKFinance Agent (v2)Finance Agent (v2)EMBEMBLegalBenchLegalBenchLegal Research BenchLegal Research BenchMedCodeMedCodeMedScribeMedScribePublic Benefits Bench v1.1Public Benefits Bench v1.1Tax Agent BenchTax Agent BenchTaxEval v2TaxEval v2
Every result as a table
TestClaude Opus 4.8GLM-5.3
Reasoning & knowledge
GPQA Diamond (Epoch AI)91.0%90.9%
LiveBench Data Analysis66.0%70.2%
LiveBench IF72.0%69.3%
LiveBench Language79.7%79.9%
LiveBench Reasoning89.2%85.8%
GPQA Diamond (Vals AI)92.4%88.1%
MMLU Pro89.6%86.8%
SimpleBench64.8%66.2%
SimpleQA Verified53.0%41.0%
Maths
FrontierMath (tiers 1-3)80.0%68.8%
AIME-style maths (OTIS mock)98.3%91.1%
LiveBench Mathematics94.3%87.9%
Chess puzzles52.0%37.0%
FrontierMath (tier 4)56.1%29.3%
Coding
Terminal-Bench 4.023.2%38.9%
LiveBench Agentic Coding50.5%60.9%
LiveBench Coding81.8%79.0%
Code Migration47.3%44.2%
LiveCodeBench87.8%80.5%
ProgramBench1.0%1.5%
SWE-bench88.6%95.4%
Vibe Code Bench v1.182.7%78.1%
DeepSWE59.0%69.0%
WeirdML82.9%75.4%
Agents
APEX-Agents48.9%56.6%
SkillsBench59.2%47.5%
Terminal-Bench-Science4.3%5.7%
DTBench94.9%87.7%
LMCA57.5%55.5%
Mystery game puzzles36.0%33.0%
Professional work
Finance Agent (v2)53.9%55.8%
EMB69.4%56.3%
LegalBench83.6%84.8%
Legal Research Bench43.8%49.0%
MedCode53.2%42.9%
MedScribe85.8%88.8%
Public Benefits Bench v1.168.1%68.5%
Tax Agent Bench30.0%43.3%
TaxEval v275.6%72.4%

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG