Bearer token authentication. Use format "Bearer your-api-key"
Successfully retrieved list of available models
Bad Request
Unauthorized
Rate Limit Exceeded
Internal Server Error
Last updated
GET /v1/models HTTP/1.1
Host: brainchat.dmind.ai
Authorization: Bearer YOUR_SECRET_TOKEN
Accept: */*
{
"data": [
{
"id": "dmind/dmind-1",
"name": "DMind: DMind-1",
"created": 1745564586,
"description": "DMind-1 is a Web3-specialized large language model(LLM) fine-tuned using a two-stage alignment process: Supervised Fine-Tuning (SFT) on 13,276 expert-labeled items from 32.7GB of curated Web3 data, followed by Reinforcement Learning from Human Feedback (RLHF). The model integrates deep knowledge across domains like DeFi, DAOs, security, and smart contracts, and supports advanced instruction following, multi-turn reasoning, and regulatory-safe outputs. RLHF is implemented via a domain-specific reward model and Proximal Policy Optimization (PPO), optimizing the model's alignment with expert preferences. Built on a state-of-the-art LLM, DMind-1 is designed for Web3-native tasks such as on-chain analysis, intelligent agent deployment, and decentralized application workflows, offering strong accuracy and interaction quality in specialized environments.",
"context_length": 32768,
"max_completion_tokens": 16384,
"quantization": "bf16",
"pricing": {
"prompt": "0.0000003",
"completion": "0.0000006",
"image": "0",
"request": "0"
}
},
{
"id": "dmind/dmind-1-mini",
"name": "DMind: DMind-1-Mini",
"created": 1745564586,
"description": "DMind-1-mini is a distilled Web3 large language model(LLM) derived from DMind-1 and a general-purpose state-of-the-art LLM via a custom multi-teacher pipeline. Trained using DMind's DeepResearch framework, it combines Web3-specific instruction distillation, distribution-level supervision, and intermediate representation alignment to preserve domain reasoning accuracy while drastically reducing inference cost. DMind-1-mini retains strong performance across tasks like agent-based interaction and DeFi analysis. With low latency and efficient compute requirements, it is optimized for real-time applications such as on-chain analytics, Q&A systems, and embedded agent deployments in decentralized environments.",
"context_length": 32768,
"max_completion_tokens": 16384,
"quantization": "bf16",
"pricing": {
"prompt": "0.0000002",
"completion": "0.0000004",
"image": "0",
"request": "0"
}
}
]
}