GPQA Diamond
?what it ishighest score on this metric 96.4%
35º / 198 · Epoch AI, latest version we checked
GLM-5.3
Benchmarks, price, context and versions
Also searched as
GLM 5, GLM 5.3, glm-5.3-flash, glm-5.3 price, glm-5.3-flashx
Direct answer: GLM-5.3 is a model from Zhipu AI (Z.ai). 90.9% on GPQA Diamond, per Epoch AI. On OpenRouter, US$ 0.07 per 1 million input tokens and US$ 1.40 per 1 million output tokens.
✓ Good for
- technical reasoning on science topics (#35 of 198 on GPQA Diamond)
- agents that follow rules and use tools (#11 of 131 on τ-bench Verified (airline))
- high-volume use (input price in the cheapest 25% on OpenRouter)
- very long documents or code (a window of 1 million tokens or more, according to OpenRouter)
- running on your own infrastructure (open weights)
✕ Not the best for
- answering short facts with little invention (#49 of 87 on SimpleQA Verified)
This block follows a rule of this section, using source data: if the model is in the top 25% of the metric, the item goes under "Good for"; if it is in the bottom half, it goes under "Not the best for". The middle band appears in neither. Price and context use OpenRouter; scores use Epoch AI or OpenRouter Benchmarks, in the latest version we checked.
Benchmarks
OTIS Mock AIME
?highest score on this metric 100%
60º / 186 · Epoch AI, latest version we checked
FrontierMath (tiers 1 to 3)
?highest score on this metric 95.1%
32º / 86 · Epoch AI, latest version we checked
SimpleQA Verified
?highest score on this metric 94%
49º / 87 · Epoch AI, latest version we checked
GPQA Diamond (OpenRouter)
?highest score on this metric 96.5%
53º / 135 · OpenRouter Benchmarks, latest version we checked
τ-bench Verified (airline)
?highest score on this metric 82%
11º / 131 · OpenRouter Benchmarks, latest version we checked
Key specs
Version history
- GLM-5.3-Flash-BF16 Hugging Face Hub, Hugging Face repo created
- GLM-5.3-BF16 Hugging Face Hub, Hugging Face repo created
- GLM-5.3 Hugging Face Hub, Hugging Face repo created
- GLM-5.3-Flash Hugging Face Hub, Hugging Face repo created
- GLM-5.2-FP8 Hugging Face Hub, Hugging Face repo created
- GLM-5.2 Hugging Face Hub, Hugging Face repo created
- GLM-5.1-FP8 Hugging Face Hub, Hugging Face repo created
- GLM-5.1 Hugging Face Hub, Hugging Face repo created
- GLM-5 Hugging Face Hub, Hugging Face repo created
- GLM-5-FP8 Hugging Face Hub, Hugging Face repo created
Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch.
FAQ
How much does GLM-5.3 cost?
On OpenRouter, US$ 0.07 per 1 million input tokens and US$ 1.40 per 1 million output tokens.
Is GLM-5.3 open source?
Yes, it has open weights. License: no data.
What is GLM-5.3?
GLM-5.3 is a model from Zhipu AI (Z.ai), with open weights, used for text, agents.
What is the context window of GLM-5.3?
1,048,576 tokens, per OpenRouter.
What is GLM-5.3's GPQA Diamond score?
90.9% on GPQA Diamond, per Epoch AI.
Is GLM-5.3 good for coding?
No data: there is no SWE-bench Verified measurement for it in open sources.
Who made GLM-5.3?
It was made by Zhipu AI (Z.ai), a lab from China.
provisional data Questions ordered by what shows up in public Google search suggestions (no volume). Answers use source data only.
Compare with
Models from Zhipu AI (Z.ai)
Sources
- Epoch AI, CC BY 4.0, latest version we checked.
- OpenRouter, dados públicos da API; exports de benchmarks e rankings em CC BY 4.0, latest version we checked.
- OpenRouter Benchmarks, CC BY 4.0, latest version we checked.
Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch. Family and version were grouped by rule from the name.