@OrlixAI just got faster.
Thanks to @AskSurplus, builders can access low-cost, high-speed inference without sacrificing performance.
Build bigger. Ship faster.
Built on @base
ORLIX AI β Platform Upgrades
β‘ Streaming Responses
No more waiting for the full reply to load. Orlix streams every token in real time you see the answer as the model thinks it, word by word. Powered by Server-Sent Events across all six providers.
πΌ Vision Upload
Drop an image, ask anything. Orlix now reads screenshots, diagrams, photos, and documents. Drag your file into the chat or click the attachment button Claude and GPT-4o will analyze it instantly.
π Real Web Search
Toggle web search on and Orlix fetches live results from Brave Search before sending your prompt to the AI. No more outdated answers the model responds with data from right now, not its training cutoff.
π§ Skill Learning Loop
Every great AI response can become a permanent skill. Save it, name it, and Orlix automatically injects it as context the next time you ask something similar. Your AI gets smarter with every conversation.
πΎ Persistent Chat History
Every conversation is saved locally and lives in your sidebar. Pick up where you left off, switch between chats, or delete what you donβt need. Your history stays on your device no cloud, no accounts.
β Multi-Model Intelligence
Seven providers. Nineteen models. One interface. Switch between Claude, Groq, GPT-4o, DeepSeek, Gemini, Grok, and Mimo without leaving the page. Find the right model for every task.
Orlix AI β Neural Command Layer. No login. No wallet. Just intelligence.
new drop for ORLIX CLI β
> memory store goals, facts, policies
> progress tracking with visual bars
> live chat 8 LLM providers, 1 key
> governance view see your rules in action
terminal-first. open source. no subscriptions.