Vascend
HorizonVascend IDEVascend CLICommunityBenchmarkContact
Sign inJoin Discord
HomeHorizonIDECLISign in
ChangelogIntelligenceCost per taskCoding Agent IndexCyber IndexDecision IndexBenchmarksSpeedLatencyPriceLeaderboard
Benchmark

GLM-5.3 (Max)

  • Z AI
  • Release · August 18, 2026
  • Open weights
  • Reasoning

Intelligence

Intelligence Index

#31 of 266
  1. 46.346MiMo-V2.6-Pro
  2. 46.346Grok 4.7 (High)
  3. 45.846GPT-6 Astra (Low)
  4. 45.445Qwen3.8 Max (0902)
  5. 45.145DeepSeek V4.1 Flash 1017 (Max)
  6. 44.845GLM-5.3 (Max)
  7. 43.944Ling 3.1 Flash 1019
  8. 43.744Step 5 Preview
  9. 43.443Claude Haiku 5.5 (Max)
  10. 42.743MiniMax-M3.1-Flash
Data from Artificial Analysis

Evaluations

  • AA-Briefcase1,505#14
  • AA-Omniscience Accuracy33.9%#67
  • AA-Omniscience Non-Hallucination70.5%#9
  • AutomationBench-AA62.2%#12
  • GDPval-AA1,650#11
  • Humanity's Last Exam42.3%#44
  • GDP.pdf11.2%#25
  • Terminal-Bench 4.041.9%#24
  • AA-LCR79.7%#51
  • SciCode59%#13
  • CritPt19.1%#38
  • Terminal-Bench-Science9.5%#14
  • FrontierCode40.1%#17
  • AA-AnalystAgent38.8%#14
  • Harvey LAB-AA0.3%#14
  • EnterpriseOps-Gym-AA36.4%#5
  • AA-3D783#17
  • ITBench-AA46.1%#6
  • MLCR-AA17.2%#14
  • PersonalityBench: Expressiveness97.5#2
  • PersonalityBench: Organization20.8#25
  • PersonalityBench: Transparency80.2#8
  • PersonalityBench: Slop30.7#16
  • PersonalityBench: Informativeness55.5#16
  • PersonalityBench: Warmth90.1#3
  • PersonalityBench: Conviction90.1#3
  • PersonalityBench: Imagination92.6#3
  • GPQA Diamond91.7%#18
  • τ²-Bench Banking50.3%#1
GLM-5.3 (Max)Median
Data from Artificial Analysis

Cost per task

Cost per task

$ per Intelligence Index task · #16 of 24
  1. $1.011.01Qwen3.8 27B (Xhigh)
  2. $1.031.03Step 5 Preview
  3. $1.131.13Mistral Large 4 Preview
  4. $1.241.24Gemini 3.8 Flash (High)
  5. $1.991.99Gemini 4 Argon (High)
  6. $2.012.01GLM-5.3 (Max)
  7. $2.112.11Kimi K3.1 Pro
  8. $3.263.26GPT-6 Astra (Max)
  9. $3.743.74Grok 4.7 (Xhigh)
  10. $4.544.54Muse Wonder (Max)
Data from Artificial Analysis

Time per task

seconds per task · #15 of 26
  1. 412412GPT-6 Astra (Max)
  2. 428428Claude Haiku 5.5 (Max)
  3. 498498Gemini 3.8 Flash (High)
  4. 507507MiniMax-M3.1-Flash
  5. 578578Mistral Large 4 Preview
  6. 604604GLM-5.3 (Max)
  7. 628628GPT-6.1 Sol (Max)
  8. 723723Gemini 4 Argon (High)
  9. 736736Claude Fable 5.1 (Max)
  10. 791791Claude Opus 5.5 (Max)
Data from Artificial Analysis

Tokens per task

output tokens per task · #19 of 27
  1. 64.3K64KMiMo-V2.6-Pro
  2. 66.8K67KQwen3.8 27B (Xhigh)
  3. 68.7K69KGLM-5.3-Flash
  4. 69.5K70KMistral Large 4 Preview
  5. 71K71KGemini 3.8 Flash (High)
  6. 71.1K71KGLM-5.3 (Max)
  7. 78.1K78KClaude Fable 5.1 (Max)
  8. 78.6K79KMiniMax-M3.1-Flash
  9. 80.6K81KGrok 4.7 (Xhigh)
  10. 89.5K90KDeepSeek V4.1 Flash 1017 (Max)
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #92 of 162
  1. 105105DeepSeek V4 Pro 0813 (Max)
  2. 103103MiniMax-M3
  3. 103103Qwen3 Omni 30B A3B (Reasoning)
  4. 100100DeepSeek V4 Pro 0813 (Non-reasoning)
  5. 9999GPT-5.3 Codex (Xhigh)
  6. 9898GLM-5.3 (Max)
  7. 9696Qwen3 Omni 30B A3B Instruct
  8. 9696Claude Opus 5.5 (Max)
  9. 9494Qwen3 Coder Next
  10. 9393Grok 4.7 (Xhigh)
Data from Artificial Analysis

Latency

seconds to first token · #96 of 162
  1. 2.32.3Qwen3.7 Plus
  2. 2.32.3Qwen3.5 122B A10B (Non-reasoning)
  3. 2.32.3Hermes 4 - Llama-3.1 405B (Reasoning)
  4. 2.32.3GPT-6 Astra (Low)
  5. 2.42.4Mistral Medium 3.5
  6. 2.42.4GLM-5.3 (Max)
  7. 2.52.5GPT-6 Luna (Low)
  8. 2.52.5Phi-4
  9. 2.62.6LongCat 2.0
  10. 2.62.6Ling 3.0 Tiny
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #120 of 169
  1. $1.761.76Inkling (Xhigh)
  2. $1.981.98DeepSeek V4 Pro 0813 (Max)
  3. $1.981.98DeepSeek V4 Pro 0813 (Non-reasoning)
  4. $22Muse Spark Open (Xhigh)
  5. $2.072.07Mistral Large 4 Preview
  6. $2.152.15GLM-5.3 (Max)
  7. $2.152.15GLM-5.3 (Low)
  8. $33Grok 4.7 (Xhigh)
  9. $33Grok 4.7 (High)
  10. $33Qwen3.8 Max (0902)
Data from Artificial Analysis

Nearby models

  • MiMo-V2.6-ProXiaomi46.3
  • Grok 4.7 (High)SpaceXAI46.3
  • GPT-6 Astra (Low)OpenAI45.8
  • Qwen3.8 Max (0902)Alibaba45.4
  • DeepSeek V4.1 Flash 1017 (Max)DeepSeek45.1
  • Ling 3.1 Flash 1019InclusionAI43.9
  • Step 5 PreviewStepFun43.7
  • Claude Haiku 5.5 (Max)Anthropic43.4
Data from Artificial Analysis

Same lab

  • GLM-5.3-FlashZ AI41.8
  • GLM-5.3 (Low)Z AI34.3
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend