1. SDK installieren
pip install openai. Der moderne v1.x-Client nutzt base_url, nicht das alte api_base.
Übergib base_url und deinen av_mcp_ Key an den offiziellen openai Python-Client, und rufe dann jedes Modell per id auf.
Der offizielle openai Python-Client nimmt eine base_url und einen api_key entgegen. Richten Sie sie auf avots und rufen Sie ein beliebiges Modell auf.
| Field | Value |
|---|---|
| Base URL | https://api.avots.ai/openai/v1 |
| API key | av_mcp_… · get your key |
| Model | anthropic/claude-sonnet-4.6 or any id from the table below |
pip install openai. Der moderne v1.x-Client nutzt base_url, nicht das alte api_base.
Erstellen Sie OpenAI(base_url="https://api.avots.ai/openai/v1", api_key="av_mcp_YOUR_KEY") oder setzen Sie OPENAI_BASE_URL und OPENAI_API_KEY in der Umgebung.
Verwenden Sie client.chat.completions.create(model="anthropic/claude-sonnet-4.6", ...). Übergeben Sie stream=True für Streaming.
from openai import OpenAI
client = OpenAI(
base_url="https://api.avots.ai/openai/v1",
api_key="av_mcp_YOUR_KEY",
)
resp = client.chat.completions.create(
model="anthropic/claude-sonnet-4.6",
messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)
Alte Anleitungen nutzen openai.api_base oder OPENAI_API_BASE aus dem 0.x-SDK. In v1.x heißen sie base_url und OPENAI_BASE_URL. Richten Sie sie immer auf den /v1 Stamm.
curl https://api.avots.ai/openai/v1/chat/completions \
-H "Authorization: Bearer av_mcp_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "anthropic/claude-sonnet-4.6",
"messages": [{"role": "user", "content": "Hello!"}]
}'
Your av_mcp_ key is not only for chat. avots also runs as an MCP server, so Python SDK can call avots creative tools, image, video, music, talking avatars, on the same key and the same balance.
| Field | Value |
|---|---|
| MCP server | https://mcp.avots.ai/ |
| API key | av_mcp_… · same key as above |
Set the model field to any id below, or a lazy name like claude / gpt / gemini that maps to a sensible default.
| Model | Best for | ⚡ / 1M tokens (in / out) |
|---|---|---|
openai/gpt-5.5-pro | Max intelligence, hardest tasks | 60 000 / 360 000 |
anthropic/claude-opus-5 | Top Claude, long reasoning | 10 000 / 50 000 |
anthropic/claude-sonnet-4.6 | Balanced (default) | 6 000 / 30 000 |
google/gemini-3.1-pro-preview | Fresh Gemini Pro | 4 000 / 24 000 |
google/gemini-2.5-pro | Stable Gemini Pro | 2 500 / 20 000 |
anthropic/claude-haiku-4.5 | Fast Claude | 2 000 / 10 000 |
deepseek/deepseek-r1 | Cheap reasoning | 1 400 / 5 000 |
google/gemini-2.5-flash | Fast and cheap | 600 / 5 000 |
openai/gpt-5-mini | Cheap OpenAI | 500 / 4 000 |
openai/gpt-oss-120b:free | Free (OpenAI open model, needs a positive balance) | Free |
* Price in ⚡ per 1,000,000 tokens, input / output (1000⚡ ≈ $1). Free models need a positive balance. Full list via GET /openai/v1/models.
Generate an av_mcp_ key in Settings, paste three fields, and reach 40+ models on one balance. The same key works for MCP.
API-Schlüssel holen ✦avots speaks the OpenAI API. Any app that talks to ChatGPT through base URL, key and model…
Add avots under Connections as an OpenAI API provider, and every avots model shows up in y…
Add one custom endpoint block to librechat.yaml and avots appears as a provider with every…
Override the OpenAI base URL in Cursor Settings, Models, paste your av_mcp_ key, add a mod…
Add an OpenAI-compatible provider to openclaw.json pointing at avots, and openclaw runs an…
Set the OpenAI base URL and your av_mcp_ key in Hermes, pick a model, and it runs on avots…
Pick OpenAI Compatible in Cline settings, paste the avots base URL and your av_mcp_ key, t…
Add one models entry to Continue config.yaml with provider openai and the avots base URL, …
Export two environment variables and launch Aider with an openai/ model prefix, and the CL…
Choose Custom (OpenAI-compatible) as the Chat Completion source in SillyTavern, paste the …
Choose Generic OpenAI as the LLM provider in AnythingLLM, fill the avots base URL, key, mo…
Pass baseURL and your av_mcp_ key to the official openai client, then call any model by id…
Add a custom OpenAI API compatible provider in Chatbox, paste the avots host and your av_m…