NewNew: 60dB Judge answers 214 of 231 JevBench decisions — 15 more than Jev, at $0.010 per 1M input tokens.See the benchmark
60db Logo

AI Chat ThatRemembers Everything

60dB VyasLanguage and reasoning model

Build intelligent chat applications with persistent conversation history, streaming responses, and memory-aware context. OpenAI-compatible API format.

Get Started FreeRead the Docs
Chat CompletionsStreamingMemory-AwareConversation HistoryAPI

Chat that does more than talk

A complete chat infrastructure for intelligent AI applications.

Chat Completions

OpenAI-compatible chat completion API. Send messages, get intelligent responses. Drop-in replacement for existing integrations.

Real-time Streaming

Stream responses token-by-token for instant feedback. Build responsive chat interfaces that feel alive.

Memory-Aware Context

Combine chat with 60db Memory for context-aware conversations. Your AI recalls past interactions and knowledge automatically.

Conversation History

Persistent conversation storage with full message history. Search, paginate, and manage chat sessions effortlessly.

Workspace Scoped

Every chat is scoped to your workspace with role-based access. Team members collaborate, data stays secure.

Simple API

Create, list, update, and delete chats with a clean REST API. Familiar format, zero learning curve.

OpenAI-compatible API

Drop-in replacement for your existing chat integrations. Same message format, same streaming protocol, with persistent history and memory built in.

  • Standard messages array format
  • Server-sent events for streaming
  • Automatic conversation storage
  • Token usage tracking per request
  • Response time metrics included
POST /v1/completions
{
  "messages": [
    {
      "role": "user",
      "content": "Summarize our last meeting"
    }
  ],
  "model": "60db-tiny",
  "stream": true
}

Power any conversation

From customer support to specialized agents, chat with memory changes everything.

Customer-Facing Chatbots

Build support bots that remember customer context across sessions. Resolve issues faster with conversation history at hand.

Educational Tutors

Create AI tutors that track student progress and adapt to their learning style over time with persistent memory.

Internal Assistants

Deploy team assistants that know your company's knowledge base. Ask questions, get answers grounded in your data.

Domain-Specific Agents

Build specialized agents for healthcare, legal, finance, or any domain. Memory ensures consistent, contextual responses.

Frequently asked questions

Yes. The chat completion endpoint uses the standard messages array format with role and content fields. If you're already using OpenAI's chat API, you can switch with minimal code changes.

Set stream: true in your request and receive server-sent events with response tokens as they're generated. This enables real-time typing effects in your chat interface.

Chat completions are billed per token at $0.50 per 1M tokens (input and output). Chat session management (creating, listing, updating, deleting chats) is completely free. You only pay for LLM inference.

Absolutely. Use the Memory API to ingest knowledge, then the Chat API references that context automatically. Your AI gives grounded responses based on your actual data.

Yes. Every chat session and its messages are automatically stored and associated with your workspace. You can retrieve full conversation history at any time via the API.

The default model is 60dB Vyas (API model ID: 60db-tiny), optimized for fast, cost-effective language and reasoning. Contact us for additional model access or specific requirements.