Model API · Available
GLM-5.3-Flash, Z.ai's thinking model, through an OpenAI-compatible endpoint on hardware AxForge owns in Stockholm. It reasons before it answers and calls your tools, and your prompts stay in the EU with zero prompt retention.
Why AxForge
| OpenAI-compatible | Chat completions, streaming, tool calls and usage reporting follow the OpenAI API shape. Point your SDK at api.axforge.ai/v1 and pick the model glm-5.3-flash. |
|---|---|
| Thinks before it answers | Every answer starts with reasoning, returned apart from the answer in reasoning_content. Good for code, analysis and multi-step agent work. |
| Zero prompt retention | Prompts and completions are processed in memory in Stockholm, Sweden. Not logged, not retained, never used for training. |
Specifications
| Model | GLM-5.3-Flash by Z.ai: mixture of experts, MIT licence |
|---|---|
| API model name | glm-5.3-flash |
| Context window | 262,144 tokens |
| Quantization | NVFP4 |
| Capabilities | Chat completions · streaming · tool calls · reasoning output on every answer · text only |
| Hardware | Two linked NVIDIA DGX Spark (GB10), owned and operated by AxForge |
| Region | Stockholm, Sweden (eu-se-1) |
| Input price | €0.35 / 1M tokens |
| Output price | €2.16 / 1M tokens |
Full details, benchmarks and FAQ on the GLM-5.3-Flash page. List prices in EUR.
How it works
| 1 | Sign in to AxForge. Your API key is created in the console. |
|---|---|
| 2 | Set base_url to https://api.axforge.ai/v1 in the OpenAI SDK you already use. |
| 3 | Send your first request with model glm-5.3-flash. |
| 4 | Watch usage in the console; every new account starts with free tokens. |
Try GLM-5.3-Flash or sign in to start.
FAQ
Yes. Chat completions, streaming, tool calls and usage reporting follow the OpenAI API shape, so the official SDKs work by changing base_url.
Yes. GLM-5.3-Flash reasons before every answer, and the reasoning counts as output tokens. For a direct answer without reasoning, use MiMo-V2.6-Flash or Qwen3.8 27B on the same key.
No. It reads text only, and a request with an image is refused with a clear message.
€0.35 per million input tokens and €2.16 per million output tokens. No subscription, no minimum.
In Stockholm, Sweden (eu-se-1), on NVIDIA DGX Spark systems AxForge owns and operates. TLS terminates in the EU.
No. Prompts and completions are processed in memory and never persisted or used for training.