GLM-5.3 API: Z.ai's reasoning model with a 1M context on FastMetal
Z.ai's GLM-5.3 is available on FastMetal as glm-5.3. We added it the day after Z.ai released it. It is the newest model in the GLM line, which already includes GLM-5.2 and is a popular family on FastMetal.
What it is for
Z.ai built GLM-5.3 as a large reasoning model for complex software engineering and agent tasks that run for a long time. Its pitch over GLM-5.2 is better coding and agent performance; those are the vendor's claims, and we have not benchmarked them here.
- A 1,048,576-token context and up to 131,072 output tokens, enough to load a whole repository in one request.
- Tool calling and structured output. We confirmed a forced tool call and strict
json_schemaoutput with real requests through FastMetal. - Text in, text out. It does not accept images.
Calling it
Set model to glm-5.3. The base URL and API key are the same as for every other model on FastMetal:
curl https://api.fastmetal.ai/v1/chat/completions \
-H "Authorization: Bearer $FASTMETAL_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "glm-5.3",
"messages": [{"role": "user", "content": "Hello"}]
}'
If you already call glm-5.2, moving up is a change to the model value and nothing else. With the OpenAI SDK, point base_url at https://api.fastmetal.ai/v1. To try it without code, pick it in chat.
The rest of the family
For volume work, GLM 5.3 Flash is the lighter model in the same family, with a longer context and a lower rate.
Pricing
Billed per million input and output tokens, with cached input at its own rate, from a prepaid balance with no monthly fee. Current rates are on pricing and the model page. Everything else we serve is in the model catalog.