Tencent Hy·HunyuanVideo

Hy-Nomas

Tencent Hunyuan’s interactive video-world model supports multi-turn creation through text instructions and camera control, generating segments while carrying scene state forward with a KV cache.

CurrentVideo generationMultimodalOpen weights

MODEL NUMBER265

First public date: 2026-10-23
Organization: Tencent Hy
Family: HunyuanVideo

Specifications

Only published specifications are shown.

Input modalities: Text↗
Output modalities: Video↗
Open weights: Yes↗
License: Apache-2.0↗
Status: Current↗

Overview

Release dates

October 23, 2026 — Tencent released Hy-Nomas code and Apache-2.0 weights on 2026-10-23 for multi-turn video-world generation controlled by text and camera paths. ↗

Why it mattered

Hy-Nomas publicly released code and Apache-2.0 weights on 2026-10-23. It extends video generation into a persistently controllable interactive world; the originator has not disclosed parameter count.

Architecture & capabilities

Model capabilities

Video generationCamera controlLong videoLocal inference

Release history

2026-10-23Open weights

Tencent released Hy-Nomas code and Apache-2.0 weights on 2026-10-23 for multi-turn video-world generation controlled by text and camera paths.

View source

Related models

Select a card to view the related model.

No related models recorded yet

Sources

Hy-Nomas: Towards Free-form Interactive World GenerationTencent Hunyuan · Published 2026-10-23 · Checked 2026-10-24

Official · Verified 02

Tencent-Hunyuan/Hy-NomasTencent Hunyuan · Checked 2026-10-24

Official · Verified 03

Tencent Hy-Nomas model card and weightsTencent (official Hugging Face organization) · Checked 2026-10-24

Official · Verified

Suggest a correction

Copy this template, add the field, proposed value, and primary source, and share it with the archive maintainer.

Model ID: hy-nomas
Page: https://llmlineage.com/en/models/hy-nomas
Field:
Proposed value:
Primary source URL:
Reason: