glm-5.3-flash API — per-token pricing and code examples
Price per 1M tokens
Cache rates apply when the provider reports cache tokens.
90% offAbout this model
glm-5.3-flash is served through OpenAI Chat Completions, OpenAI Responses, Anthropic Messages. You pay $0.015 per 1M input tokens and $0.05 per 1M output tokens. That is 90% below the official price of $0.15 and $0.50. Cache reads cost $0.003 per 1M tokens when the provider reports cached tokens. It is included in budget packs from $1.99 up.
What GLM-5.3-Flash is good at
Z.ai introduces GLM-5.3-Flash as the first native multimodal model in the GLM-5 series, released on August 26, 2026. Where GLM-5.3 reads text only, the Flash model takes video, images, text and files as input and answers in text. Z.ai lists a 1M-token context and a maximum output of 128K tokens; in the API reference, max_tokens goes up to 131072 for this series.
The use cases Z.ai names are visual and practical: turning screenshots into UI code, working with office documents such as PPTX, PDF, DOCX and XLSX, finance workflows, video understanding, 3D work in Blender, Godot projects, CAD and computer use. If your agent has to look at a design mockup, a spreadsheet export or a screen recording before it acts, GLM-5.3-Flash is the GLM model built for that input.
Forced thinking and preserved reasoning
On GLM-5.3-Flash, thinking cannot be switched off. Z.ai's API reference says the series can only be enabled and that depth is controlled by reasoning_effort, and the thinking-mode guide describes it as forced thinking. For this model reasoning_effort supports low, high and max, with max as the default.
The second setting worth knowing is thinking.clear_thinking. In the API reference it defaults to true. Setting it to false keeps earlier reasoning across turns, which Z.ai calls preserved thinking, and the GLM-5.3-Flash model page recommends false. The same page suggests temperature 1, top_p 0.95 and reasoning_effort max as starting values.
For tools, Z.ai's reference allows up to 128 functions per request. If you want tool arguments to stream in as they are generated, tool_stream defaults to false and only works together with stream: true.
{
"model": "glm-5.3-flash",
"reasoning_effort": "high",
"thinking": { "type": "enabled", "clear_thinking": false },
"messages": [{ "role": "user", "content": "Turn this screenshot into a React component" }]
}
This example is in the /v1/chat/completions format, which is the format Z.ai's API reference documents for these fields.
What it costs on Cent AI
GLM-5.3-Flash is charged per token from your prepaid USD balance, at the four rates in the price box on this page. Because thinking is forced and defaults to max, every request carries reasoning tokens, billed as output. Lowering the effort to low is the main lever on that part of the bill. Preserved thinking also has a cost side: keeping earlier reasoning in the conversation means more tokens are sent with each later turn.
Z.ai documents context caching as implicit, with no manual setup, and reports hits in usage.prompt_tokens_details.cached_tokens. Z.ai's cache guide does not state a cache lifetime or a minimum prompt length. Cent AI charges the cache read rate only when cached tokens appear in that usage field.
Using it from your tools
GLM-5.3-Flash is offered on /v1/chat/completions, /v1/responses and /v1/messages. Cent AI does not change your parameters silently, so the value you choose for thinking, clear_thinking or reasoning_effort is not replaced with a different one. Keep thinking.type at enabled in every request, since Z.ai documents no other mode for GLM-5.3-Flash, and remove any old setting that tries to disable thinking before you switch an existing GLM integration to this model.
Example: the cost of one request
Take a request that sends 2,000 new input tokens, reads 8,000 tokens from cache and gets 800 output tokens back. At the rates in the price box on this page:
- input: 2,000 × $0.015 / 1,000,000 = $0.00003
- cache read: 8,000 × $0.003 / 1,000,000 = $0.000024
- output: 800 × $0.05 / 1,000,000 = $0.00004
That request costs $0.000094, under Rp2 at Cent AI's exchange rate on 2026-11-14 (Rp18,231.96 per USD). A thousand requests like it cost $0.094. Your real requests will differ; each one shows its own amount in usage.cost and on the Requests page.
Questions
Is there a 5-hour or weekly limit? No. You pay per token from your balance. Each key accepts up to 600 requests per minute, and you can set your own daily or monthly limit on a key.
Is a failed request charged? A request that fails before the model receives it costs $0. Tokens the model already produced are charged.
Does unused balance expire? No, not while your account exists. Every top-up is manual, with no auto-renewal.
Related models
Compare with glm-5.3, deepseek-v4.1-flash, minimax-m3.1-flash, or see every model on the pricing page.
Specification
- Context window
- —
- Input and output
- Text
- Endpoints
- /v1/chat/completions, /v1/responses, /v1/messages
- Smallest pack
- $1.99
Latency and error rate per model are on the status page. status.cent-ai.dev
What a request costs
Open the full calculatorCost per request
Estimate
glm-5.3-flashAn estimate at price version 4. Your bill comes from the usage the provider reports for each request.
Save $0.0419Code
curl https://api.cent-ai.dev/v1/chat/completions \
-H "Authorization: Bearer $CENT_AI_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "glm-5.3-flash", "messages": [{"role": "user", "content": "Hello"}]}'Set CENT_AI_KEY to a key from the dashboard. OpenAI SDKs use https://api.cent-ai.dev/v1; Anthropic SDKs use https://api.cent-ai.dev.
Price history
Version 4, in effect since Nov 14, 2026.
Effective
per 1M tokens
Inputold $0.015 new $0.015Cache readold $0.0015 new $0.003Cache writeold $0.015 new $0.015Outputold $0.05 new $0.05Effective
per 1M tokens
Inputold $0.0075 new $0.015Cache readold $0.00075 new $0.0015Cache writeold $0.0075 new $0.015Outputold $0.025 new $0.05Effective
per 1M tokens
Inputold $0.0075 new $0.0075Cache readold $0.0015 new $0.00075Cache writeold $0.0075 new $0.0075Outputold $0.025 new $0.025
Compare with
- deepseek-v4.1-flashInput$0.015
- gpt-6-lunaInput$0.01978
- qwen3.8-flashInput$0.015
- deepseek-v4-flashInput$0.015
Questions
How much does glm-5.3-flash cost on Cent AI?
$0.015 per 1M input tokens and $0.05 per 1M output tokens. Cache reads cost $0.003 and cache writes $0.015 per 1M tokens. Each request is rounded up once.
Is there a minimum charge?
A successful request costs at least $0.00002. A request that ends early, because the client left or the output was cut off, pays only for its tokens.
How do I pay?
Top up a prepaid USD balance with QRIS or USDT/USDC, or buy a budget pack once. Nothing renews automatically.
What if the provider does not report usage?
The request is billed from an estimate and marked estimated on the Requests page. An estimate never contains cache tokens.
Which endpoint do I call?
/v1/chat/completions, /v1/responses, /v1/messages. The OpenAI base URL is https://api.cent-ai.dev/v1.
Try glm-5.3-flash with one key
One prepaid balance for every model. Pay with QRIS or USDT/USDC.