Let me conclude first: When accessing domestic LLM, do not guess the model ID based on the manufacturer name. The correct process is to first query the real-time model list, then use the real model ID to complete the minimum request, and finally verify business capabilities such as streaming output, tool invocation, image or retrieval item by item.
Check before you start
Names such as DeepSeek, Tongyi Qianwen, Kimi, Doubao, and GLM usually represent manufacturers or model series, and are not necessarily equal to the model ID that can be directly used by the API. The availability, price, context, and capabilities of different models will vary, and this article does not provide a fixed list.
Prepare the following information:
- API Key created by the console.
- Base URL
https://www.aifast.club/v1。 /v1/modelsOr the actual model ID displayed by the console.- Business must-have capabilities, such as streaming output, tool invocation, visual understanding, or retrieval.
Configuration steps
1. Get the real-time model list
curl https://www.aifast.club/v1/models \
-H "Authorization: Bearer $AIFAST_API_KEY"Copy the model ID from the response. Don't pass family names like "DeepSeek" "Kimi" or "GLM" directly as request parameters unless the real-time API actually returns the exact same ID.
2. Complete minimal text call
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["AIFAST_API_KEY"],
base_url="https://www.aifast.club/v1",
)
response = client.chat.completions.create(
model="YOUR_MODEL_ID",
messages=[{"role": "user", "content": "Explain your main abilities in three points。"}],
)
print(response.choices[0].message.content)Only after the minimum text call is successful, continue to add streaming output, structured output, tools or multi-modal parameters.
3. Choose models based on capabilities rather than manufacturer name
| demand | What to verify first |
|---|---|
| Chinese writing and summarizing | Instruction following, factual accuracy, long text stability |
| Code and agent tasks | Tool calls, structured output, context and error recovery |
| Image understanding | Whether the API accepts image input and its format restrictions |
| Search questions and answers | Whether it has search capabilities, citation sources and timeliness boundaries |
| Batch tasks | rate limiting, concurrency, failure retry and complete task cost |
4. Preserve switchable configurations
Do not scatter model IDs through application code. Store the Base URL, API Key, and model ID in environment variables or a configuration service so staged rollouts and rollback remain manageable.
FAQ
Why can the console see the model, but the call prompts "model not found"?
Common reasons are that the display title is copied instead of the real model ID, the model is not yet open to the current account, or the client splices the wrong path. Request first /v1/models And check the final URL.
Can domestic models share OpenAI SDK?
When the target API supports the required OpenAI Compatible request format, you can first verify the underlying call with the OpenAI SDK. Specific capabilities still need to be tested item by item, and they cannot be completely consistent by default.
How to control multi-model switching costs?
Record input and output tokens, number of retries, delays and task success rates. For models that are cheap but require many retries, the full mission cost may be higher.
Next step
Continue reading How to choose text, image, video, and retrieval models and LLM API cost estimation and usage control, establish verifiable model selection and budgeting rules.