Chat with Orion – a visual agent that sees, reasons and acts
Hey HN! We’re excited to share Orion [1] — our new visual agent that sees, reasons, and acts across images, videos, and documents. Frontier VLMs (GPT, Claude, Gemini) can describe what they see, but they can’t reliably act on visual inputs. Ask them to detect objects, segment images, or chain visual steps — they’ll fail in surprisingly inconsistent ways. High-res images collapse to ~1024px. And the visual AI ecosystem is fragmented across separate APIs for image understanding, OCR, image-gen, video-gen, etc. We built Orion to fix this. Orion combines VLM reasoning with reliable…
In plain words
Orion is a visual agent that processes images, videos, and documents to see, reason, and act on visual information. Unlike general AI models that struggle with visual tasks like object detection and image segmentation, Orion combines visual language model reasoning with reliable computer vision tools in a unified interface. It handles high-resolution images, chains multiple visual steps together, and consolidates what is typically fragmented across separate APIs for image understanding, OCR, and generation into one platform.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hey HN! We’re excited to share Orion [1] — our new visual agent that sees, reasons, and acts across images, videos, and documents. Frontier VLMs (GPT, Claude, Gemini) can describe what they see, but they can’t reliably act on visual inputs. Ask them to detect objects, segment images, or chain visual steps — they’ll fail in surprisingly inconsistent ways. High-res images collapse to ~1024px. And the visual AI ecosystem is fragmented across separate APIs for image understanding, OCR, image-gen, video-gen, etc. We built Orion to fix this. Orion combines VLM reasoning with reliable computer-vision tools inside a unified chat-completions interface. You can chain visual steps, inspect results, and treat visual tasks the same way you treat text workflows. Here’s a quick demo [2]. What Orion can do today: - Detect objects, faces, people (with precise, visualized boxes) - Segment objects or salient regions interactively - Edit, remix, and re-imagine images/videos from prompts - Summarize visual content (images or videos) - Transform images: crop, rotate, upscale - Transform videos: trim, sample, highlight scenes - Parse and structure documents: pagination, layout, OCR, extraction One unified “chat-completions”-like interface — no juggling multiple vision APIs. Check out the tours in the chat [3] or read the announcement [4]. API access opens next week. Happy to answer any questions — otherwise, feel free to try the tours and break things! [1] Learn more about Orion: https://vlm.run/orion [2] Promo video: https://youtu.be/cPJN4iZz6QQ [3] Chat: https://chat.vlm.run [4] LinkedIn announcement: https://www.linkedin.com/posts/sudeeppillai_ai-computervisio...
Does the same job
all alternatives →
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
Launched alongside, November 2025
the whole month →
- IB
Life & fun · Nov 2025 · bitsnpieces.dev
- BBoing▲782
Life & fun · Nov 2025 · boing.greg.technology