OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Evaluation & Observability

Vllora

Vllora is a self-hostable Rust server and web UI for tracing and debugging AI-agent interactions routed through an OpenAI-compatible chat-completions gateway.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Vllora — OpenViST project profile in Agent Evaluation & Observability.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Vllora when developing or debugging an agent and you want model requests and tool activity routed through a gateway with traces available in a local UI.

Who it is for

  • Developers investigating model calls, tool use and workflow behavior in AI agents.
  • Teams integrating agent workflows with LangChain, Google ADK, OpenAI-compatible requests or MCP servers.

Documented uses

  • Route agent chat-completion requests through a local gateway and capture interaction traces.
  • Inspect agent calls, tool interactions and workflow activity in real time.
  • Connect MCP servers over HTTP and SSE for external tool access.
Before you choose

Requirements & limits

The documented quick start installs with Homebrew and runs a local server and UI. The README describes the source as always visible and identifies the Vllora license as Elastic License 2.0; the inner llm package has a separate Apache 2.0 license.

  • Existing agents need to send model requests through Vllora's OpenAI-compatible chat-completions endpoint to have those interactions traced.
  • The documented Homebrew setup requires Homebrew; provider API keys are configured through the local UI.
  • MCP server connections are documented over HTTP and SSE.
  • The Vllora application and its inner llm package have different licenses.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install and start the local Vllora server

Install Vllora with Homebrew and open its local UI before routing requests through the gateway.

Before you start

  • Homebrew installed

Steps

  1. Run brew tap vllora/vllora.

  2. Run brew install vllora.

  3. Run vllora to start the server.

  4. Open http://localhost:9091 for the UI; the gateway listens at http://localhost:9090.

Expected outcome

A local Vllora server and UI are running, ready for provider API-key configuration and gateway requests.

Guide limitations

  • This setup sequence does not itself configure provider credentials or demonstrate a completed agent trace.
Documentation 1 ↗
Source-based practical guide

Send a chat-completions request through Vllora

Configure a provider key in the UI and submit the documented sample request to the local gateway.

Before you start

  • A running Vllora server
  • An AI provider API key configured through the UI at http://localhost:9091

Steps

  1. Visit http://localhost:9091 and configure the AI provider API key.

  2. Send a JSON request to http://localhost:9090/v1/chat/completions with a model and messages, as shown in the repository README.

  3. Inspect the interaction's trace and debugging information in the Vllora UI.

Expected outcome

A chat-completions interaction passes through the OpenAI-compatible gateway, where Vllora collects trace and debugging information.

Guide limitations

  • The documented request uses the gpt-4o-mini model; provider access and credentials are needed for a request to work.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
vllora/vllora · Official homepage title · retained in its source language
How we review sourcesSuggest a correction