Search the full model catalog.

Tencent HunyuanVideo-1.5-i2v animates one reference image into a prompt-guided video clip with coherent motion, 480p–1080p presets, and reproducible seeds.

SenseNova U1 8B Visual Understanding creates infographic-style images and prompt-based edits from text and an optional reference image. Built by SenseNova on NEO-Unify.

Netflix VOID Model removes objects from videos and rewrites shadows, reflections, and motion after the edit. Use it for VFX cleanup and counterfactual scene edits.

Tencent’s HunyuanVideo 1.5 T2V turns detailed prompts into MP4 video clips with coherent motion. Control resolution, aspect ratio, frames, steps, and seed.

Qwen3.6-27B is a 27B dense vision-language model from Qwen for agentic coding and reasoning, with optional thinking traces in outputs. On Wiro a request takes about 25K input tokens and returns up to 16K.

A real-time, multilingual text-to-speech system offering expressive voice design and high-fidelity voice cloning through low-latency streaming inference.

Reconstruct a 3D scene from multi-view photos or a short video. Produces 3D Gaussian splats with depth maps and cameras, plus a colored point cloud.

Multilingual speech-to-text for 25 European languages with auto language detection, punctuation, capitalization, and optional timestamps.

A 14B autoregressive video model from PKU-YuanGroup, fine-tuned from Wan 2.1, that covers text-to-video, image-to-video and video-to-video and builds clips up to a minute long in 33-frame chunks.

OmniVoice by k2-fsa generates 24 kHz speech from text in 600+ languages. Clone a speaker from a short reference clip or design a new voice from attributes.

Decensored Qwen3.5-4B checkpoint for chat, coding, and analysis. Supports optional thinking traces and sampling controls for output style. On Wiro a request takes about 72K input tokens and returns up to 16K.

A large language model optimized for chat interactions and logical reasoning tasks. Designed for developers and researchers.

Qwen3.5-4B is the 4B model of Qwen's Qwen3.5 line. A Thinking field turns its reasoning pass on or off, and Wiro gives it room for 72,704 input tokens per request.

A dense 9-billion parameter language model optimized for chat and reasoning tasks. Designed for efficient deployment and high-quality responses.

CohereLabs cohere-transcribe-03-2026 is a 2B Conformer speech-to-text model for 14 languages. It creates accurate transcripts for meetings, calls, and audio archives.

A large language model optimized for chat interactions and reasoning tasks, designed for advanced users seeking powerful natural language processing capabilities.

Combines separate photos of people and pets into one posed picture. Pick one of 22 scenes, from oil-painted pet portraits to black-and-white family shots, and upload the subjects it asks for.

Song Generation 2 generates full songs with vocals and instrumentals from lyrics — supports 14 genres, reference audio cloning, and separate track output (vocal/bgm/mix).

Multilingual text-to-speech from Hume AI that clones a voice from a reference clip, built on Llama 3.2 3B. Its tokenizer pairs each text token with one acoustic vector, which keeps the audio aligned to the script.

Writes an instrumental track for a video and mixes it with the clip's own audio. Describe the music in a caption, or leave it empty and the app picks a style from the footage.