Tasks and inputs
Hy-Nomas is listed for Text-to-video, Video-to-video. It accepts text and produces video. Use the publisher’s input format and preprocessing for your chosen deployment.
Hy-Nomas is Tencent Hunyuan's open interactive world model: it generates 720p video segment by segment, steered by text and camera controls.
The specification score summarizes recorded features and limits. It is not a benchmark result.
Pricing conditions: No API pricing stated; open weights for self-hosting
Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.
Accepts
Produces
Imagine directing a video while it plays. Add a lantern. Make it rain. Pan left and walk through the door. Hy-Nomas is built for that kind of session: it generates a world one video segment at a time and takes new text instructions and camera trajectories between segments.
Tencent Hunyuan released it on 21 October 2026, with a Diffusers build three days later and a web demo.
Diffusion happens in a compressed latent space; autoregression happens across segments.
| Setting | Value |
|---|---|
| Denoising steps (distilled) | 8 per segment |
| Output | 720×1280 at 24 FPS |
| Standard config | Sequence parallel across GPUs |
| Single-GPU config | One RTX 4090: INT8 DiT, INT8 KV cache, FP8 decoder, CPU offload |
| Licence | Apache-2.0 |
The single-4090 path is the notable one. Interactive world models usually assume a datacenter.
Tencent also released Hy-Nomas-Bench: four suites, 280 test cases, covering free-form multi-turn text interaction, camera navigation, mixed text and camera control, and spatial memory. It includes keyboard timelines and reference camera paths, which makes it handy for evaluating any interactive video model, not just this one.
It is a different tool from prompt-to-clip generators in the text-to-video directory. Those produce a finished shot; Hy-Nomas produces an ongoing, steerable scene. Other Tencent releases are collected on Tencent's provider page.
Script a ten-turn session that mixes object insertion, a style change and a camera loop that returns to the start. The return is the test: does the room look the same when you come back? Run it once on the multi-GPU config and once on a single 4090 to see what the quantised path costs in quality.
Compare the model on the work you need to complete.
Hy-Nomas is listed for Text-to-video, Video-to-video. It accepts text and produces video. Use the publisher’s input format and preprocessing for your chosen deployment.
Choose duration, resolution and frame rate together. Compare the cost of an accepted clip, including repeat generations.
The weights are downloadable under Apache-2.0. Check runtime support and size the complete model and working context for your expected concurrency.
These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.
Genie 4 is Google DeepMind's interactive world model with real-time native audio, available in Project Genie to eligible Google AI Ultra subscribers.
Apple
FastFrames-L is Apple's open 14B autoregressive two-step video diffusion model that generates 480p video at interactive latency, distilled from Wan2.1-14B.
Gemini Omni 1.2 Flash is Google's conversational video generation and editing model, making 3-15 second clips with native audio from 360p to 4K.
pika
Pika H3 Flex is Pika's open LoRA post-training of MiniMax H3: six-step sampling, up to 15 s clips with 32 kHz stereo audio at 480p or 768p, 24 fps.
heygen
HeyGen Video 1 generates 5 to 30 second MP4 clips with audio from text, images or reference media, up to 2K at 24 fps. Built on MiniMax H3.
Meta
Muse Video is Meta’s previewed video-generation model with native audio; a public Meta Model API route and price have not been established.