How to Use GLM 5.3 Flash on Composite

At a glance

GLM-5.3-Flash is a native multimodal model from Z.ai, described as suited for efficient coding and long-horizon agent tasks. Z.ai credits its hybrid sparse and linear attention design with keeping long-context behavior accurate. On Composite it takes text, images and video, and has a window of just over a million tokens.

It always thinks

Unlike most models on the free list, GLM 5.3 Flash cannot answer without reasoning first. You choose how hard it thinks, not whether. The lowest setting is low, so for chat where you want the reply to start quickly, name it in the model ID: z-ai/glm-5.3-flash:free[low]. Use [high] or [max] for multi-step problems.

Many chat apps can show or hide the model's reasoning. If your app prints the thinking into the chat, look for its reasoning or "show thoughts" setting. A [none] suffix does not exist, and asking this model for no reasoning will not turn it off.

Images and video

GLM 5.3 Flash reads video as well as images. Most roleplay frontends only send text and pictures, so video matters mainly through the API, for example asking the model to describe or summarize a clip. Replies are text only.

For coding and agents

The model supports tool calling with parallel calls and structured output, which covers what coding assistants and agent frameworks expect from an OpenAI-compatible endpoint. Point the tool at https://composite.lucidity.sh/v1 with the free ID above. Agent loops make many calls per task, and every call counts toward the 80 a day, so a long agent session can use the allowance quickly. The paid ID, z-ai/glm-5.3-flash, has no daily count.

Connect it to your chat app

1. Create an API key on the API Keys page.
2. In SillyTavern, ChubAI, Agnai or RisuAI, choose an OpenAI-compatible provider and set the base URL to https://composite.lucidity.sh/v1. JanitorAI wants the full URL: https://composite.lucidity.sh/v1/chat/completions.
3. Paste the key, then type z-ai/glm-5.3-flash:free as the model name.
4. Send a short message to check the connection before you load a long chat.

Step-by-step setup for each app is in the JanitorAI and SillyTavern guides, and the docs hub covers the others.

From code, it is a normal OpenAI-compatible request:

curl https://composite.lucidity.sh/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "z-ai/glm-5.3-flash:free", "messages": [{"role": "user", "content": "Hello"}]}'

Free access (as of October 21, 2026)

On October 21, 2026, this model is on Composite's free list as z-ai/glm-5.3-flash:free. The plain z-ai/glm-5.3-flash stays paid. Every account shares 80 requests a day across all of the free models, separate from Synth's count, and the count resets at midnight UTC. Credits and plans do not raise it.

The free list changes. Check the free models on the Models page for the current set, and the features guide for the rules. A :free id that is no longer on the list returns a model-not-found error and charges nothing. When the 80 requests are used up, requests stop with an error until the reset; Composite never charges your credits to continue. You can switch to z-ai/glm-5.3-flash on credits or a plan, or to Synth.

More

• Model page with live pricing and rankings
• Model modifiers, including reasoning effort suffixes
• Best LLMs for Roleplay in 2026
• All model guides