πŸ€– UVR LLM Dashboard

Enter password to access

Wrong password

UVR LLM Dashboard

DGX Spark
πŸ”‘ For API key access, visit: wiki.uvrlab.org/arrc:account:ai_tools

🌐 Web Chat (Open WebUI)

https://everyone-herself-short-monkey.trycloudflare.com

πŸ“‘ API Endpoint

πŸ€– Available Models

πŸ–₯️ Cursor IDE Setup

  1. Open Cursor Settings β†’ Models β†’ OpenAI API Key section
  2. Set Base URL to: https://extracellularly-unsoulish-adriel.ngrok-free.dev/v1Set API Key to your personal token
  3. Set Model name to: minimax-m2.7
  4. Click "Verify" to test the connection

Supports both chat and agent mode (tool calling enabled).

πŸ’» cURL Example

curl https://extracellularly-unsoulish-adriel.ngrok-free.dev/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-m2.7",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": true
  }'

🐍 Python (OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    base_url="https://extracellularly-unsoulish-adriel.ngrok-free.dev/v1",
    api_key="YOUR_API_KEY"
)

response = client.chat.completions.create(
    model="minimax-m2.7",
    messages=[{"role": "user", "content": "Hello!"}],
    stream=True
)
for chunk in response:
    print(chunk.choices[0].delta.content or "", end="")
minimax-m2.7 πŸ†•
MiniMax-M2.7 230B MoE AWQ-4bit Β· TP=2 Β· ~38 tok/s Β· πŸ”§ tools Β· πŸ’­ reasoning Β· 128K ctx Β· native streaming
gemma4-26b
Gemma 4 26B MoE NVFP4 Β· ~51 tok/s Β· πŸ–ΌοΈ vision Β· πŸ”§ tools
Select a model and start chatting. Supports image (πŸ“·) and audio (🎀) uploads for vision/audio models.