Artificials
Compare models
All makers

AI model maker

Z.ai

37 models on Artificials, 14 with published test results and 29 with open weights you can download.

RSS feed
Best on the capability index155.6GLM-5.3, #27 of 274
Best rankTop 3%GLM-5.3-Flash on Arena math
Newest releaseGLM 5.3 PrimeSep 23, 2026
Models3714 tested, 29 with open weights

Where Z.ai stands

Its best model’s rank among all models tested on each skill area’s main test, the same tests model pages use. A longer bar means a better rank.

  • OverallGLM-5.3 on Epoch Capabilities Index
    Top 10%
  • CodingGLM-5.2 on LiveBench coding
    Top 32%
  • MathGLM-5.3-Flash (max) on Mock AIME 2024–2025
    Top 21%
  • KnowledgeGLM-5.3 (max) on SimpleQA Verified
    #43 of 77
  • AgentsGLM-5.3 on LiveBench agentic coding
    Top 23%
  • LanguageGLM-5.3 on LiveBench language
    Top 47%
  • Data analysisGLM-5.3-Flash on LiveBench data analysis
    Top 50%
  • ReasoningGLM-5.2 (max) on GPQA Diamond
    Top 12%

On the capability index

Its 6 models on the Epoch Capabilities Index, each with its rank among all 274 models. The top score is 167.3, held by Claude Opus 5.5 from Anthropic.

Every test

Its best result on each published test, and that result’s rank among every model tested. Scores from different tests are never combined.

TestBest modelScoreRank
Epoch Capabilities IndexGLM-5.3155.6#27 of 274
Arena mathGLM-5.3-Flash1503#11 of 372
Arena hard promptsGLM-5.3 (max)1503#21 of 389
Arena instruction followingGLM-5.3 (max)1476#21 of 389
Arena textGLM-5.3 (max)1479#24 of 389
Arena codingGLM-5.3-Flash1523#25 of 384
Arena creative writingGLM-5.2 (max)1454#26 of 387
GPQA DiamondGLM-5.2 (max)91.9%#25 of 222
Arena WebDevGLM-5.3 (max)1623#17 of 123
SWE-bench VerifiedGLM-5.2 (max)78.7%#5 of 32
SWE-bench Verified, bash onlyGLM-5 (high)72.8%#7 of 42
Mock AIME 2024–2025GLM-5.3-Flash (max)93.9%#41 of 198
LiveBench agentic codingGLM-5.360.9%#14 of 63
DeepSWEGLM-5.3 (max)69.0%#7 of 26
FrontierMath Tiers 1–3GLM-5.3 (max)68.8%#24 of 81
LiveBench codingGLM-5.279.7%#20 of 63
Chess puzzlesGLM-5.2 (max)21.0%#55 of 141
CursorBenchGLM-5.3 (max)42.6%#6 of 14
FrontierMath Tier 4GLM-5.2 (max)29.3%#29 of 63
LiveBench languageGLM-5.379.9%#29 of 63
LiveBench overallGLM-5.376.1%#30 of 63
LiveBench data analysisGLM-5.3-Flash76.4%#31 of 63
LiveBench instructionsGLM-5.369.3%#32 of 63
LiveBench mathematicsGLM-5.289.8%#32 of 63
LiveBench reasoningGLM-5.385.8%#32 of 63
SimpleQA VerifiedGLM-5.3 (max)41.0%#43 of 77

Every model

Newest first. The price is per 1M tokens, blended three input tokens to one output token: Z.ai’s own listing, or else the middle price across hosts.

ModelReleasedContextPriceCapability index
GLM 5.3 PrimeSep 23, 20261M tokens$4.30Not on the index
GLM-5.3-FlashXSep 18, 20261M tokens$0.59Not on the index
GLM-5.3-FlashAug 26, 20261M tokens$0.238151.9
GLM Flash LatestAug 26, 20261M tokens$0.152Not on the index
GLM-5.3Aug 14, 20261M tokens$2.15155.6
GLM 5.3 FastAug 14, 20261M tokens$3.23Not on the index
GLM-5.3Aug 14, 20261M tokens$2.15Not on the index
GLM-5.3 HighspeedAug 14, 20261M tokensNo paid listingNot on the index
GLM-5.2Jun 13, 20261M tokens$2.15151.8
GLM 5.2 FastJun 13, 20261M tokens$3.23Not on the index
GLM-5.2Jun 13, 20261M tokens$2.15Not on the index
GLM-5.2 (fp8)Jun 13, 20261M tokens$1.50Not on the index
GLM LatestMay 3, 20261M tokens$3.00Not on the index
GLM-5.1Apr 7, 2026200K tokens$2.15149.8
GLM-5V-TurboApr 1, 2026200K tokens$1.90Not on the index

Models, prices, context windows and release dates from models.dev (MIT). Test results from LiveBench (Apache-2.0), Epoch AI (CC BY 4.0), Datacurve DeepSWE, compiled by Epoch AI (CC BY 4.0), Cursor CursorBench, compiled by Epoch AI (CC BY 4.0), Arena (CC BY 4.0) and SWE-bench (CC BY-NC 4.0). A model counts here when Z.ai sells it, several hosts list it or a test covers it. Each figure is the source’s own; nothing is combined into a new score.