discord Registration
BlazeAPI Complete Guide: Model List and Integration Instructions
Platform Overview
BlazeAPI provides a high-performance model integration environment, supporting basic text and image generation, as well as advanced video and music generation models.
- API Base URL:
https://blazeai.boxu.dev/api/(Note: no need to add/v1at the end) - Authentication: Login authorization via Discord
Public Model Catalog
Below are the main models currently supported by the platform, along with their health status and usage costs:
| Model Name | Status (Health) | Supported Features | Usage Cost |
|---|---|---|---|
| MiniMax-Hailuo-2.3-Fast |  Running Normally |
Video Generation / Streaming | 5x req/use |
| MiniMax-M2.1 / M2.5 / M2.7 | Running Normally |
Role Play (RP) / Streaming | 1x req/use |
| MiniMax-image-01 | Running Normally |
Image Generation / Streaming | 5x req/use |
| MiniMax-music-2.5 / 2.6 | Running Normally |
Music Generation / Streaming | 5x req/use |
| MiniMax-speech-2.8-hd | Running Normally |
Speech Synthesis (TTS) | 5x req/use |
| DeepSeek-V4 (Flash/Pro) |  Slight Delay |
Streaming / Role Play | 1x req/use |
| Qwen 3.5/3.6 Omni Series | Running Normally |
Vision / Search / Chain-of-Thought / Video | 1x req/use |
| GLM 4.7 / 5.1 | Running Normally |
Tool Calling (Tools) / Streaming | 1x req/use |
Note: Models with the “Thinking” suffix support logical reasoning; models with the “Search” suffix support online search.
Quick Start Guide
1. Get Your API Key
After logging in to the BlazeAPI Dashboard, you can find your personal API key starting with blz_.
2. Free Quota
- Daily request limit: 1,000
- Requests per minute (RPM): 20
- Reset time: Automatically resets at a fixed time every day
3. Code Integration (Python Example)
Since the API is compatible with the OpenAI format, you can use the official openai library directly:
from openai import OpenAI
# Initialize the client
client = OpenAI(
api_key="your_API_KEY",
base_url="https://blazeai.boxu.dev/api/"
)
# Send a chat request
response = client.chat.completions.create(
model="qwen3.6-plus", # You can also choose other models
messages=[{"role": "user", "content": "Hello, please introduce yourself."}]
)
print(response.choices[0].message.content)
Notes
- Base URL path: Make sure to use
/api/at the end, not the standard OpenAI/v1/. - Model status:
Healthy: Available for normal use.Degraded: May have slow responses.Dead: Model is temporarily offline.
- Multimodal Support: The Qwen and MiniMax series provide extensive vision, image, and video processing capabilities. Please refer to the specific model documentation when calling.
Start your AI development journey now!
