ModelBeat
Models/GLM-5.3
All models
Z

Z.ai (Zhipu AI): GLM-5.3

z-ai/glm-5.3

GLM-5.3 is Z.ai (Zhipu AI)’s open-weights AI model, released Aug 14, 2026. It scores 157 on the Epoch Capabilities Index, costs $1.40/M input and $4.40/M output tokens, and handles a 1M-token context window. Its strongest use case is coding, where it ranks in the 88th percentile of the models Model Beat tracks.

GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks.

Context
1M
Input
$1.40 / 1M
Output
$4.40 / 1M
Cheapest via SiliconFlow
$0.70 / $1.90 / 1M
Released
Aug 14, 2026
Modalities
text→text
Strongest atCoding top 12%Agentic & Tools top 30%Reasoning & Knowledge top 39%

Benchmarks

Scores on standardized evaluations. Higher is better — and rank shows where GLM-5.3 lands among all models tracked on Model Beat.

63.0
Intelligence Index
Epoch AI
63rd percentile of tracked models
88.0
Coding Index
Epoch AI
88th percentile of tracked models
70.0
Agentic Index
Epoch AI
70th percentile of tracked models

Reasoning

5 evals
GPQA Diamond90.9%

Graduate-level scientific reasoning

Humanity's Last Exam42.3%

Frontier of human expert knowledge

SimpleBench66.2%

Common-sense trick questions

SimpleQA Verified41.0%

Factual accuracy & hallucination

WeirdML75.4%

Novel ML problem-solving

Coding

2 evals
SciCode59.0%

Scientific research coding

WebDev Arena1614

Human-rated web development

Math

1 evals
AIME 2024/202591.1%

Olympiad-qualifier math

Agentic & Tools

1 evals
APEX56.6%

Multi-step agentic tasks

In the news

16

Changelog

20
  • GLM-5.3: cheapest credible provider via SiliconFlow now $1.9 per 1M output tokens (was $2.2, -14%)cheapest provider · Nov 20, 2026
  • GLM-5.3: cheapest credible provider via Cloudflare now $2.2 per 1M output tokens (was $2, +10%)cheapest provider · Nov 18, 2026
  • GLM-5.3: cheapest credible provider via Cloudflare now $0.7 per 1M input tokens (was $0.6, +17%)cheapest provider · Nov 18, 2026
  • GLM-5.3: cheapest credible provider via Tenstorrent now $0.6 per 1M input tokens (was $0.643, -7%)cheapest provider · Nov 16, 2026
  • GLM-5.3: cheapest credible provider via Morph now $0.643 per 1M input tokens (was $0.4, +61%)cheapest provider · Nov 13, 2026
  • GLM-5.3: cheapest credible provider via Morph now $1.978 per 1M output tokens (was $2.464, -20%)cheapest provider · Nov 13, 2026
  • GLM-5.3: cheapest credible provider via PrimeIntellect now $2.464 per 1M output tokens (was $2.2, +12%)cheapest provider · Nov 10, 2026
  • GLM-5.3: cheapest credible provider via PrimeIntellect now $0.4 per 1M input tokens (was $0.7, -43%)cheapest provider · Nov 10, 2026
  • GLM-5.3: cheapest credible provider via Cloudflare now $2.2 per 1M output tokens (was $1.4, +57%)cheapest provider · Nov 6, 2026
  • GLM-5.3: cheapest credible provider via Cloudflare now $0.7 per 1M input tokens (was $0.086, +714%)cheapest provider · Nov 6, 2026
  • GLM-5.3: cheapest credible provider via Morph now $0.086 per 1M input tokens (was $0.226, -62%)cheapest provider · Oct 31, 2026
  • GLM-5.3: cheapest credible provider via Morph now $1.4 per 1M output tokens (was $2.208, -37%)cheapest provider · Oct 31, 2026
  • GLM-5.3: cheapest credible provider via Morph now $2.208 per 1M output tokens (was $2, +10%)cheapest provider · Oct 29, 2026
  • GLM-5.3: cheapest credible provider via Morph now $0.226 per 1M input tokens (was $0.5, -55%)cheapest provider · Oct 29, 2026
  • GLM-5.3: cheapest credible provider via Sail Research now $2 per 1M output tokens (was $1, +100%)cheapest provider · Oct 23, 2026
  • GLM-5.3: cheapest credible provider via Sail Research now $0.5 per 1M input tokens (was $0.75, -33%)cheapest provider · Oct 23, 2026
  • GLM-5.3: cheapest credible provider via Sail Research now $1 per 1M output tokens (was $1.505, -34%)cheapest provider · Oct 21, 2026
  • GLM-5.3: cheapest credible provider via Sail Research now $0.75 per 1M input tokens (was $0.5, +50%)cheapest provider · Oct 21, 2026
  • GLM-5.3: cheapest credible provider via Relace now $0.06 per 1M input tokens (was $0.95, -94%)cheapest provider · Oct 4, 2026
  • GLM-5.3: context window changed from 1,310,720 to 1,048,576 tokenscontext · Sep 30, 2026

Compare GLM-5.3 with

Frequently asked questions

What is GLM-5.3?

GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks.

Who created GLM-5.3?

GLM-5.3 is an AI model developed by Z.ai (Zhipu AI), released Aug 14, 2026.

How much does GLM-5.3 cost?

GLM-5.3's list price from Z.ai (Zhipu AI) is $1.40 per million input tokens and $4.40 per million output tokens. The cheapest credible third-party provider on OpenRouter (SiliconFlow) serves it at $0.70/$1.90 per 1M.

What is GLM-5.3's context window?

GLM-5.3 supports a context window of up to 1M tokens.

How does GLM-5.3 perform on benchmarks?

On standardized evaluations tracked by Epoch AI, GLM-5.3 scores 59.0% on SciCode, 1614 on WebDev Arena, 42.3% on Humanity's Last Exam.

Is GLM-5.3 open source?

GLM-5.3 is released with open weights (Open weights (restricted use)), so the model can be downloaded and self-hosted.

Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Nov 21, 2026.