GLM 5.2: What It Does and What It Costs
z-ai/glm-5.2Z.ai’s open-weight reasoning model for long text and project-level software work.
1.05M tokens of context — about 786,432 words, on a scale that ends at 2M.
What it can do
- Reasoning
- Tool use
- Open weights
- Takes text
- Gives text
Released 16 Jun 2026 · checked 3 Oct 2026.
About it
GLM 5.2 is a text-in, text-out model from Z.ai with a context window of 1,048,576 tokens. It is suited to long-running agent workflows and project-level software engineering. Its input price is $0.41 per million tokens and its output price is $3.99 per million tokens. Open weights are available.
Review
In a line
GLM 5.2 is a text-based reasoning model from Z.ai. It can take in and produce text, and its weights are open.
What it's good at
The model is suited to long-running agent workflows and software engineering across a project. It supports reasoning and tools, and can handle large amounts of text at once.
Price in plain terms
Input costs $0.41 per million tokens; output costs $3.99 per million tokens. Tokens are small pieces of text that the model reads or writes. Reading a 500-word page costs about 0.027 cents. A chat turn with about 1,000 input tokens and 500 output tokens costs about 0.24 cents; 1,000 such turns cost about $2.40.
Context and limits
GLM 5.2 can take in up to 1,048,576 tokens at once—about a million. It can produce up to 943,718 tokens in one answer.
Ways to use it
Its model ID is z-ai/glm-5.2. Its weights are open, and its Hugging Face ID is zai-org/GLM-5.2.
Verdict
Consider GLM 5.2 if you need a model for long text, reasoning, or project-level software work. Its input price is low compared with its output price, so tasks that generate a lot of text can cost more. It is not a fit for image tasks, since it works with text only.
Verdict
GLM 5.2 is worth considering for people who need a text model that can take in very large documents or support extended, tool-using workflows. Its open weights may also matter to buyers choosing among models. Budget for output: a million generated tokens costs $3.99, nearly ten times the input rate. Choose another model if your work depends on images; GLM 5.2 handles text only.
- Open weights are available.
- Handles up to 1,048,576 tokens of text at once.
- Supports reasoning and tools.
- Designed for long-running agent workflows and project-level software work.
- Text only; no image input or output.
- Output tokens cost more than input tokens.
- Pick it ifLong-text tasks and extended agent workflows; Project-level software engineering
- Skip it ifYou need a model that can take images as input or produce them.
Pricing
Input is $0.41 per million tokens and output is $3.99 per million. The example costs are about 0.027 cents to read a page, 0.24 cents for one typical chat turn, and $2.40 for 1,000 such turns.
Questions
Who makes GLM 5.2?
Z.ai makes GLM 5.2.
Can GLM 5.2 work with images?
No. Its inputs and outputs are text.
How much text can it take in at once?
Its context window is 1,048,576 tokens, or about a million tokens.
How much does it cost to generate text?
Output costs $3.99 per million tokens.
How much is 1,000 chat turns?
About $2.40 for 1,000 turns with roughly 1,000 input tokens and 500 output tokens each.
Are GLM 5.2's weights open?
Yes. Its Hugging Face ID is zai-org/GLM-5.2.
On these boards
More from Z.ai
- GLM 5.3 Prime$2.80 in
- GLM 5.3 FlashX$0.37 in
- GLM Flash Latest$0.026 in
- GLM 5.3 Flash$0.15 in
- GLM 5.3 Flash (batch)$0.060 in
- GLM Latest$0.12 in
- GLM 5.3$1.40 in
- GLM 5.3 (batch)$0.45 in
- GLM 5.1$0.96 in
- GLM 5V Turbo$1.20 in
- GLM 5 Turbo$1.20 in
- GLM 5$0.60 in
- GLM 4.7 Flash$0.060 in
- GLM 4.7$0.60 in
- GLM 4.6V$0.30 in
- GLM 4.6$0.43 in
- GLM 4.5V$0.60 in
- GLM 4.5$0.60 in