Hotshot – 4 Person Team Builds a State of the Art Video Model
Hi HN! We're proud to share Hotshot, a large-scale diffusion transformer model for text-to-video generation that we built with just a 4-person team. You can try it today in beta at https://hotshot.co, with 2 free generations per day. The model generates 5 seconds of 720p video from text prompts. It excels at prompt alignment, and consistency. It also excels at generating people, animals, and nature. In blind tests with 100 users, Hotshot generations were preferred to Runway ML 60% of the time. Hotshot generations were preferred to Luma 80% of the time. Overall, users preferred…
In plain words
Hotshot is a text-to-video generation model that creates 5-second videos in 720p from text prompts. Built by a 4-person team, it uses a diffusion transformer approach trained on 600 million video clips and 1 billion images. The model emphasizes prompt alignment, consistency, and realistic generation of people, animals, and nature scenes. Users can try it in beta at hotshot.co with 2 free generations daily. It is designed for anyone looking to create video content from text descriptions.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hi HN! We're proud to share Hotshot, a large-scale diffusion transformer model for text-to-video generation that we built with just a 4-person team. You can try it today in beta at https://hotshot.co, with 2 free generations per day. The model generates 5 seconds of 720p video from text prompts. It excels at prompt alignment, and consistency. It also excels at generating people, animals, and nature. In blind tests with 100 users, Hotshot generations were preferred to Runway ML 60% of the time. Hotshot generations were preferred to Luma 80% of the time. Overall, users preferred Hotshot's results to other publicly available text-to-video models ~70% of the time. We built this model from scratch with a 4 person team in just 4 months. It is trained on 600 million video clips and 1 billion images. It uses a custom-trained video captioner for better temporal understanding and a custom autoencoder for efficient long sequence training. We've detailed more technical aspects of the journey in a blog post: https://hotshot.co/release Some technical highlights include A. scaling to thousands of GPUs, tackling infrastructure and optimization challenges. B. developing custom kernels and data parallelism techniques. C. Creating a Watchdog system to detect and respond to GPU process hangs. D. Optimizing data streaming and compression for efficient training. We believe that this model is just the beginning. In the next 12 months, entire YouTube videos will be AI generated by creators. Text to video models like this one lay the foundation for this and much more. Control over every aspect of generations, longer durations, higher resolutions, real time interactivity, and more modalities (like audio!) are just around the corner. We're here to answer any questions about the model, our training process, or our plans for the future. We're also always looking for talented individuals to join our team! We'd love for you to try our 2 free generations per day and let us know what you think. We're excited to see what the HN community will create with it!
Does the same job
all alternatives →
- BABuild an open-source computer vision model in seconds using text2023 · usezeroshot.com · ▲64
Hello HN! I want to share something me and a few friends have been working on for a while now — Zeroshot, a web tool that builds image classifiers using text-image models and autolabeling. What does this mean in practice? You can put together an image classifier in about 30 seconds that’s faster and more accurate than CLIP, but that you can deploy yourself however you’d like. It’s open source, commercially licensed, and doesn’t require you to pay anyone per API call. Here's a 2 minute video that shows it off: https://www.youtube.com/watch?v=S4R1gtmM-Lo How/why does it…
- LSLemon Slice Live – Have a video call with a transformer model2025 · ▲195
Hey HN, this is Lina, Andrew, and Sidney from Lemon Slice. We’ve trained a custom diffusion transformer (DiT) model that achieves video streaming at 25fps and wrapped it into a demo that allows anyone to turn a photo into a real-time, talking avatar. Here’s an example conversation from co-founder Andrew: https://www.youtube.com/watch?v=CeYp5xQMFZY. Try it for yourself at: https://lemonslice.com/live. (Btw, we used to be called Infinity AI and did a Show HN under that name last year: https://news.ycombinator.com/item?id=41467704.) Unlike existing…
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
Launched alongside, August 2024
the whole month →
- IY
Life & fun · 2024 · ytch.xyz
- IA
Hey there HN! We’re Joe and Stopa, and today we’re open sourcing InstantDB, a client-side database that makes it easy to build real-time and collaborative apps like Notion and Figma. Building modern apps these days involves a lot of schleps. For a basic CRUD app you need to spin up servers, wire up endpoints, integrate auth, add permissions, and then marshal data from the backend to the frontend and back again. If you want to deliver a buttery smooth user experience, you’ll need to add optimistic updates and rollbacks. We do these steps over and over for every feature we build, which can…
Dev tools · 2024 · github.com