priorsBuilder
Support

Settings

Theme

Explore themes

Mode

Library

Looks borrowed from benchmark sites and products. Themes with a light and a dark version follow your mode.

Bring your own theme

Upload a Priors theme file, or have your agent make one: copy the prompt, tell it the look you want, and upload what it returns. With the Priors skill installed it already knows the format.

Custom themes stay in this browser. They travel inside shared chart links and a page's theme, so others see them as you do.

For the builder, Explore and model pages on this device. Published pages keep the theme their owner picked.

Mode

Zhipu · released 2026-08-18

GLM 5.3

Measured on 196 benchmarks across 2 reasoning efforts.

31.5

#24 of 66 models · best at max

Reasoning efforts in this ranking

max31.5#58 of 149
low23.1#96 of 149

Strengths

  • GLM 5.3
  • Gemini 4 Carbon (#1)
  • Field median
bestField medianFrontierCode 2.0-…Terminal-Bench v4…AA-Briefcase v1.1AutomationBench-AATerminal-Bench-Sc…HLE-DiamondCritPt v2 Hard Su…AA-Omniscience v1…AA-Omniscience v1…GDP.pdf
  1. FrontierCode 2.0-AA15%21.6%#28 of 66
  2. Terminal-Bench v4-AA10%41.9%#14 of 66
  3. AA-Briefcase v1.115%1503#21 of 66
  4. AutomationBench-AA5%62.2%#19 of 66
  5. Terminal-Bench-Science-AA15%9.5%#24 of 66
  6. HLE-Diamond10%17.9%#36 of 66
  7. CritPt v2 Hard Subset10%33.5%#27 of 66
  8. AA-Omniscience v1.1 Accuracy10%32.3%#38 of 66
  9. AA-Omniscience v1.1 Non-hallucination5%61.1%#32 of 66
  10. GDP.pdf5%11.2%#44 of 66

Bars show where GLM 5.3 sits among the 66 ranked models on each benchmark (right is best); the tick is the field median.

Cost and speed

against 66 ranked models

Cost per task

#35 cheapest of 48

$0.52at max

Time per task

#41 quickest of 48

30 minat max

Output speed

#43 fastest of 57

65 tok/sat max

Time to first token

#23 quickest of 57

3.6sat max

Output tokens per task

#40 fewest of 52

119kat max

Token price

$0.309 per 1M tokens at API prices$0.032 on GLM Coding Pro (新客 ¥538) 闲时

97.0% cache read · 2.50% fresh input · 0.00% cache write · 0.50% output

Assumed mix: no measured T3 Code usage for this model.

This ranking’s benchmarks

10 of 10 measured

agentic

AutomationBench-AA

244 results
  1. #34max of 24462.2%
  2. #88low of 24452.0%

Leader: Gemini 4 Carbon · 77.8%

agentic

AA-Briefcase v1.1

249 results
  1. #40max of 2491503
  2. #108low of 2491185

Leader: Sonnet 5.5 · 1813

reasoning

CritPt v2 Hard Subset

114 results
  1. #57max of 11433.5%
  2. #80low of 11421.5%

Leader: GPT-6 Astra · 72.0%

coding

FrontierCode 2.0-AA

112 results
  1. #53max of 11221.6%
  2. #74low of 11212.7%

Leader: Opus 5.5 · 46.4%

economics

GDP.pdf

242 results
  1. #136max of 24211.2%
  2. #157low of 2428.8%

Leader: Muse Wonder Full · 32.6%

reasoning

HLE-Diamond

151 results
  1. #73max of 15117.9%
  2. #74low of 15117.5%

Leader: Gemini 4 Carbon · 69.1%

knowledge

AA-Omniscience v1.1 Accuracy

120 results
  1. #78max of 12032.3%
  2. #79low of 12031.9%

Leader: Fable 5.1 · 68.4%

knowledge

AA-Omniscience v1.1 Non-hallucination

120 results
  1. #41max of 12061.1%
  2. #77low of 12036.8%

Leader: Gemini 4 Carbon · 95.8%

reasoning

Terminal-Bench-Science-AA

125 results
  1. #57max of 1259.5%
  2. #76low of 1253.8%

Leader: Gemini 4 Carbon · 64.8%

coding

Terminal-Bench v4-AA

240 results
  1. #36max of 24041.9%
  2. #44low of 24034.8%

Leader: Kimi K3.1 · 67.2%

All 196 benchmarks

show 186 more