GLM 5.3 Flashx
GLM-5.3-FlashX is the high-speed variant of Z.ai's GLM-5.3-Flash, a native multimodal model delivering inference speeds of up to 200 tokens/s.
glm-5.3-flashxReleased 18 Sep 2026 (Sep 2026)Pricing
List prices at Tencent TokenHub, in US dollars per million tokens.
| Tier | Input / 1M | Output / 1M | |
|---|---|---|---|
| Standard | $0.37 | $1.25 | |
| Cached input 80% off input | $0.075 | 80% off input | |
| Cache write, 5 minutes | $0.00 |
What it costs a month
At Tencent TokenHub list prices, 30 days.
- Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$178
- RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$152
- Coding agent1,000 requests a day, 40K in / 2K out, cached$271
- Overnight batch20,000 requests a day, 4K in / 0.4K out$1,188
Where to run it
2 hosts serve GLM 5.3 Flashx. The comparison is on blended price, 3 parts input to 1 part output.
| Host | In / out per 1M | Model id | |
|---|---|---|---|
BigModel (Zhipu AI)Cloud glm-5.3-flashx | $0.286 / $1.00 −21% vs headline | glm-5.3-flashx | |
Tencent TokenHubCloud glm-5.3-flashx | $0.37 / $1.25 | glm-5.3-flashx |
Capabilities
Core
- Streaming
- Function calling
- JSON mode
- Structured outputs
Multimodal
- Vision
- Image input
- Audio
- Image generation
- Video
Advanced
- Reasoning
- Tool calling
- MCP
- Prompt caching
- Batch API
- Embeddings
- Takes in
- TextImagesVideo
- Produces
- Text
- Features
- Prompt cachingVideo input
Cheaper alternatives
Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.
Same model, cheaper host
- BigModel (Zhipu AI)21% lessglm-5.3-flashx$0.286 / $1.00
Smaller Zhipu AI models
No cheaper Zhipu AI model keeps its key capabilities.
Other frontier labs
- 22% less
- 66% less
- 66% less
- 56% less
Open-weight picks
- 11% less
- 18% less
- 26% less
- 41% less
Model line
Newer and older versions, at each lab's own price.
- Sep 2026GLM 5.3 FlashxLatest$0.37 / $1.25
- UndatedGLM 4.7 Flashx$0.07 / $0.40
Call it
Using Tencent TokenHub's API.
from litellm import completion
response = completion(
model="tencent/glm-5.3-flashx",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Quick answers
How much does GLM 5.3 Flashx cost?
On Tencent TokenHub, GLM 5.3 Flashx costs $0.37 per million input tokens and $1.25 per million output tokens. Cached input costs $0.075 per million tokens. The cheapest host is BigModel (Zhipu AI), at $0.286 per million input tokens and $1.00 per million output tokens.
Which APIs offer GLM 5.3 Flashx?
GLM 5.3 Flashx runs on 2 APIs: BigModel (Zhipu AI) and Tencent TokenHub.
What is GLM 5.3 Flashx's context window?
1M tokens, with up to 128K tokens in one reply.
Is GLM 5.3 Flashx being retired?
No retirement date is scheduled for it on any API we track.
Is there a newer version of GLM 5.3 Flashx?
No. GLM 5.3 Flashx is the newest model in its line.
What are cheaper alternatives to GLM 5.3 Flashx?
Models that cost less and keep most of what it can do: GPT-5.4 Nano on OpenAI (22% less), Claude Haiku 5.5 on Anthropic (66% less) and DeepSeek Flash on DeepSeek (11% less). GLM 5.3 Flashx itself costs 21% less on BigModel (Zhipu AI). Savings are on a mix of 3 input tokens to 1 output token.
When was GLM 5.3 Flashx released?
Zhipu AI released it on 18 Sep 2026.
What is GLM 5.3 Flashx's API model id?
Use glm-5.3-flashx with Tencent TokenHub.
Sources
Raw catalog entry for glm-5.3-flashx on Tencent TokenHub
{
"cache_creation_input_token_cost": 0,
"cache_read_input_token_cost": 7.5e-8,
"input_cost_per_token": 3.7e-7,
"input_cost_per_token_cache_hit": 7.5e-8,
"litellm_provider": "tencent",
"max_input_tokens": 1000000,
"max_output_tokens": 128000,
"mode": "chat",
"output_cost_per_token": 0.00000125,
"source": "https://www.tencentcloud.com/document/product/1300/78937",
"supported_endpoints": [
"/v1/chat/completions"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_video_input": true,
"supports_vision": true
}