Tencent-Hunyuan/Prism
Can I trust it?
Be careful with this one.
Stars from real accounts No data
We haven't checked its stars for fakes yet.
Active OK
Last change 5 days ago.
Licence Careful
No licence: legally you can't reuse the code, unless the README says otherwise.
Quick checks from public data. They help, but they are no guarantee.
What it is
Research code for Prism, a dynamic sparse attention framework for natively training joint video-audio generation models at 2K resolution. It partitions tokens into spatiotemporal zones and adapts block shapes and sparsity per query, using video feature variance and…
Who it's for
Researchers and ML engineers training high-resolution joint video-audio diffusion models in Python on CUDA GPUs
How to use it
Ask your AI to set it up
Paste this into Claude, ChatGPT or your coding assistant.
Help me install and set up Tencent-Hunyuan/Prism in my project. Follow its README: https://github.com/Tencent-Hunyuan/Prism. Explain each step and check it is safe before running anything.
Star curve
Get alerts for repos like this
We'll email you when something in Models & inference starts rising. Free, once a week, only what matches.
18,263 GitHub repos watched and ranked every hour. Figures as of 20 October 2026.
What arrives: one email on Mondays with the repos that match, never more. See last week's. Privacy
For maintainers
Add this badge to your README. It shows the repo's place on the radar, updates every hour and links back to this page.
[](https://repoalerts.com/repo/Tencent-Hunyuan/Prism?ref=badge)Similar repos
- Tencent-Hunyuan/Hy-Memory
Python SDK and memory engine for LLM agents with three scenes: hierarchical long-term memory for chat, fact and experience memory for working tasks, and diff-chain version tracking for coding. Support
- code-video-model/CodeVideoModel
A research codebase for generating videos from editable Three.js scenes. It produces Three.js code from text prompts, renders proxy videos, creates reference images, and runs the MiniMax-H3 model to p
- Hongyang-Du/PaperLeet
An offline practice library for turning research papers into code. It offers 49 papers and 63 exercises covering attention, reinforcement learning, diffusion and flow, and related methods. Each exerci
- j1s4nn/jev-adaptive-granularity-WM
Official code and audit artifacts for routing per-segment fine, coarse, and hold temporal granularity in the Causal-rCM (Wan2.1-1.3B) causal video generator, using a text-only Jev oracle to pick modes
- zjwfufu/SMD
A training and inference codebase for few-step diffusion distillation using an on-policy method. It provides configurations for distilling text-to-image, image editing, text-to-video, audio-video and
- mistralai/timestral
Standalone inference code for Timestral, a family of time series foundation models for multivariate probabilistic forecasting. Supports past-only, known-future and categorical covariates, outputs 21 q