SAM 3 is a unified foundation model for promptable segmentation in images and videos. It can detect, segment, and track objects using text or visual prompts such as points, boxes, and masks.

Moondream2 is a highly efficient open-source vision language model that combines powerful image understanding capabilities with a remarkably small footprint.

Sa2VA is an MLLM capable of question answering, visual prompt understanding, and dense object segmentation at both image and video levels

Build real-time voice applications powered by Grok. Send recorded audio and get a voice and text response.

Marlin is a 2B video VLM tuned for the two questions developers actually want to ask of their videos: what is happening, and when?

MoonDreamNext Batch is a multimodal vision-language model for batch captioning.

Sa2VA is an MLLM capable of question answering, visual prompt understanding, and dense object segmentation at both image and video levels

Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks

Speak in one language, hear it in another. Streaming speech-to-speech translation with the translated text alongside the audio.

Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks

Explore interactive worlds in real time with movement and camera controls.

Explore interactive worlds in real time with movement and camera controls.

Direct continuous, realtime video streams with live prompts while preserving characters, settings, and story continuity.

Direct a continuous world in real time with image references and text prompts.

OpenAI spec compatible endpoint of Isaac-01 which is a multimodal vision-language model from Perceptron for various vision language tasks.

Real-time games rendered using VAST's Eden Live 14B

Real-time games rendered using VAST's Eden Live 5B