The Model Index
Up to four models

GPT-5.4 vs GLM-5

On the index (GPT-4 = 100), GPT-5.4 scores 166 and GLM-5 142. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

GPT-5.4GLM-5
Next to GPT-5.4
Presets
GPT-5.4
166150–182
WeightsClosed
ReleasedMar 2026
Price in / out$2.5 / $15
Context1.05M
Core tests taken7 of 9
GLM-5
142126–157
WeightsOpen
ReleasedFeb 2026
Price in / out$1.0 / $3.2
Context205K
Core tests taken3 of 9

GPT-5.4 leads GLM-5 by 24 points, ahead on 3 of 3 shared core tests, at 3.6× the price.

Test by testthird-party results

GPT-5.4GLM-5
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)GPQA Diamond (Vals AI)GPQA Diamond (Vals AI)MMLU ProMMLU ProCL-benchCL-benchMATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)FrontierMath (2025 set)FrontierMath (2025 set)FrontierMath tier 4 (2025 set)FrontierMath tier 4 (2025 set)CODINGSWE-bench VerifiedSWE-bench VerifiedLiveCodeBenchLiveCodeBenchSWE-benchSWE-benchVibe Code Bench v1.1Vibe Code Bench v1.1WeirdMLWeirdMLPROFESSIONAL WORKCorpFin v2CorpFin v2LegalBenchLegalBenchTaxEval v2TaxEval v2NOVEL PROBLEMSARC-AGI-2ARC-AGI-2ARC-AGI-1ARC-AGI-1
Every result as a table
TestGPT-5.4GLM-5
Reasoning & knowledge
GPQA Diamond (Epoch AI)93.3%87.8%
GPQA Diamond (Vals AI)91.7%83.3%
MMLU Pro87.5%86.0%
CL-bench27.9%18.7%
Maths
AIME-style maths (OTIS mock)97.8%80.0%
FrontierMath (2025 set)47.6%16.4%
FrontierMath tier 4 (2025 set)27.1%2.1%
Coding
SWE-bench Verified76.9%72.1%
LiveCodeBench84.1%81.9%
SWE-bench78.2%71.4%
Vibe Code Bench v1.167.4%23.4%
WeirdML77.7%48.2%
Professional work
CorpFin v265.3%62.9%
LegalBench86.0%84.1%
TaxEval v274.0%70.0%
Novel problems
ARC-AGI-274.0%4.9%
ARC-AGI-193.7%44.7%
Task length · METR45 sec

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

This page as a chartDownload PNG