Z.ai
GLM 5.3 Flash
GLM 5.3 Flash is an open-weight mixture-of-experts (MoE) model from Z.ai with 320 billion total parameters, 18 billion of them active, and a 1M-token context window. It is the first natively multimodal model in the GLM-5 series, built on a newly trained base model with a hybrid of sparse and linear attention to cut long-context serving costs. It is released under the MIT license.
- Model ID
- zai-org/GLM-5.3-Flash
GLM 5.3 Flash specifications
| Developer | Z.ai |
|---|---|
| Architecture | Mixture of experts (MoE) with hybrid sparse and linear attention |
| Total parameters | 320 billion |
| Activated parameters | 18 billion per token |
| Experts | 288 routed experts, 8 selected per token, 1 shared |
| Layers | 45 |
| Context length | 1,048,576 tokens (1M) |
| Native input modalities | Text and images |
| Languages | English and Chinese |
| Pre-training data | 30 trillion tokens of multimodal data |
| Released weights | FP8 |
| Reasoning | Adjustable thinking budget: low, high and max (default max) |
| License | MIT |
Built for
- Coding and agentic tasks at a low cost per token
- Image and text understanding in one model
- Long-context work across up to 1M tokens
On Alchemy Inference
Supported features
- Streaming
- Tool calling
- Structured outputs
- JSON mode
- Reasoning
- Prompt caching
Input
Text and images
API endpoints
Chat completions, Completions, Responses (OpenAI-compatible)
Frequently asked questions about GLM 5.3 Flash
GLM 5.3 Flash is Z.ai's mixture-of-experts model with 320 billion total and 18 billion active parameters, image and text input and a 1M-token context window. It is the first natively multimodal model in the GLM-5 series and is released under the MIT license.
GLM 5.3 Flash starts from a newly trained base model, accepts images as well as text, and activates 18 billion parameters per token, with a hybrid of sparse and linear attention for cheaper long-context serving. GLM 5.3 is the larger, text-only model built on GLM-5.2's base. Z.ai reports that GLM 5.3 Flash outperforms GLM-5.2 at one-tenth of its price.
Yes. GLM 5.3 Flash natively accepts text and images, and on Alchemy you can send both.
Yes. Z.ai documents a thinking budget with three levels, low, high and max, and the model defaults to max when no level is set.
The MIT license, according to its Hugging Face model card.
With Token Based APIs, call GLM 5.3 Flash through Alchemy's OpenAI-compatible API with the model ID zai-org/GLM-5.3-Flash and pay for the tokens you use. Dedicated Deployments are not offered for this model. Join the waitlist for early access; once you're in, the in-app quickstart walks you through your first request.
Current pricing and limits for GLM 5.3 Flash are listed in your dashboard once you're in.
Alchemy Inference operates with zero data retention (ZDR).
Start building with GLM 5.3 Flash.
Join the waitlist for early access and up to $1,000 in free credits.