Skip to content
Yardstick
  1. Models
  2. Z.ai
  3. GLM-5.3

GLM-5.3

Z.ai · Flagship · Aug 18, 2026NewOpen weights

Z.ai's open-weight flagship for long-running engineering work, and the strongest open model on Terminal-Bench 4.0 in this catalogue.

Announcement (opens in a new tab)
Artificial Analysis Intelligence Index
44.8#11 of 29 · max effort
API price per 1M tokens
$1.40 / $4.40input / output · $2.15 blended
Context window
1Mup to 128K output tokens
Output speed
79 t/s28s to first answer at max effort

Benchmark results

Each row shows this model (blue) against every other model in the catalogue (grey), worst on the left, best on the right.

Overall capability

  • Artificial Analysis Intelligence Index44.8#11 of 29

    Max effort

    11.357.6
  • Epoch Capabilities Index155.6#11 of 16
    146.9167.3
  • LiveBench76.1%#14 of 23
    63.9%83.4%

Coding

  • Terminal-Bench 4.041.8%#6 of 13

    Claude Code harness, max effort

    14.6%64.8%
  • SWE-Bench Pro (V2 Hard)84.3%#4 of 5

    Mini-swe-agent harness, max effort

    58.8%92.2%
  • Arena WebDev1,622#13 of 27

    Max effort

    1,2631,814

Agents & professional work

  • GDPval-AA1,651#8 of 18

    Max effort

    1,2421,866
  • Vending-Bench 2$8,164#5 of 9
    $2,158$15,515

Reasoning

  • FrontierMath (Tiers 1–3)68.8%#10 of 14

    Max effort

    53%93.7%
  • SimpleBench66.2%#10 of 12
    45.8%88.4%

Knowledge & science

  • Humanity's Last Exam (Diamond)16.4%#8 of 9
    13.4%60.6%
  • GPQA Diamond90.9%#10 of 14

    Max effort

    90.2%95.8%

Human preference

  • Arena Text1,478#11 of 24

    Max effort

    1,4251,525

Reasoning effort trade-offs

The same model at different thinking budgets. More effort buys a higher intelligence score at a higher cost per task and a longer wait. Measured by Artificial Analysis.

EffortScoreCost / taskFirst answerSpeed
Low
34.3
$0.8528s80 t/s
Max (headline score)
44.8
$2.0128s79 t/s

Specs & pricing

Developer
Z.ai
Released
Aug 18, 2026
Availability
Generally available
API model ID
glm-5.3
Weights
Open · Custom licence
Parameters
753B
Reasoning
Always on
Accepts
Text
Produces
Text
Context window
1,000,000 tokens
Max output
128,000 tokens
Input price
$1.40 / 1M tokens
Output price
$4.40 / 1M tokens
Cached input
$0.26 / 1M tokens

Closest alternatives

Models from other labs nearest to GLM-5.3 on Artificial Analysis Intelligence Index.

  • Qwen3.8-Max

    Alibaba

    Intelligence 45.4 · $2 / $6

    Compare side by side

  • Kimi K3

    Moonshot AI

    Intelligence 43.6 · $3 / $15

    Compare side by side

  • Claude Haiku 5.5

    Anthropic

    Intelligence 43.4 · $0.10 / $0.50

    Compare side by side

Sources

  • Announcement (opens in a new tab)
  • Model documentation (opens in a new tab)
  • Pricing (opens in a new tab)
Yardstick

A plain reading of how today’s SI models compare. Scores are taken from each benchmark’s own leaderboard wherever one exists; anything a lab reported about its own model is marked.

Data as of Oct 8, 2026.

Explore

  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs
  • Methodology & sources

Sources

  • Artificial Analysis (opens in a new tab)
  • Arena (opens in a new tab)
  • Epoch AI Benchmarking Hub (opens in a new tab)
  • ARC Prize (opens in a new tab)
  • Terminal-Bench (opens in a new tab)
  • Scale AI Labs (opens in a new tab)

Yardstick is an independent reference and is not affiliated with, sponsored by or endorsed by any lab listed. Product names and logos are trademarks of their respective owners and are used only to identify them. Benchmarks are imperfect proxies: check a model on your own task before you commit to it.

  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs
  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs
  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs