Reliable workflows for verified free AI models

OpenConvo

OpenConvo turns volatile free AI models into a useful local-first workspace: task presets, fallback routing, file context, web research, projects, memory, and self-hosted control.

OpenConvo
Workflow ready
Compare these notes and draft a launch plan.
openai/gpt-oss-20b:free
I’ll treat this as a launch workflow: compare the notes, extract risks, and produce a practical plan with sources and next actions.

Built for messy real-world AI work.

Free models can be rate-limited, inconsistent, and hard to trust. OpenConvo adds workflow intent, safer routing, and local control around that messy reality.

Reliable free-model routing

Use verified-free OpenRouter models with cooldowns, fallback handling, and clear shared limits.

Task-first workflows

Switch from generic chat to quick answers, research, file analysis, code help, or writing support.

Self-host anywhere

Clone the MIT-licensed app, deploy it yourself, and keep control of your workspace.

Task presets for research, files, code, writing, and quick answers
Dynamic verified-free OpenRouter model list
Streaming chat with fallback handling
File-aware conversations and PDF context
Optional web search and research mode
Projects, prompts, memory, and artifacts
Local-first data with import and export

Open-source first, hosted for convenience.

The hosted website should make OpenConvo easy to try. The source code should remain the foundation: transparent, self-hostable, and friendly to people who want their own keys and their own data.

Open-source promise

OpenConvo is MIT-licensed, self-hostable, and built around local browser storage so users can inspect the code, bring their own keys, and keep control of their workspace.

Use the hosted app for a quick start, or bring your own keys when you want more reliable personal capacity.

Ready to try the workspace?

Use the hosted app now, or self-host it with your own keys and deployment limits.

Open app