HEhexfaker·May 27, 2026Submission
AVTR-1: Open-weight real-time flow-matching transformer for audio-driven avatars
2 points · Open discussionavaturn-live
avtr-1
Python2026–2026
avaturn-live / avtr-1
avtr-1
AVTR-1: Avatars that listen back
2Distinct people
3Mentions
0Explicit recommenders
549GitHub stars
What to know before exploring
What it is
AVTR-1: Avatars that listen back
GitHub
Where it is used
No documented product use has been verified yet.
What people are saying
No recommendation phrase was found in the indexed comments.
No classified warning is attached to this page. Read the discussions before treating silence as approval.
Also in these discussions
No other repository is repeatedly linked in the same discussions.
This week on Hacker News
Nov 7–Nov 14 UTC · Direct repository links and separately labeled related project stories. Indexed through 11/14/2026, 5:04:57 PM UTC.
No direct repository links were found in the indexed HN discussions over the past seven days. Older discussions are available below.
Mentions over time
HN mentions by hour3 dated mentions
Observed mentions: 2026-05-26 15:59:18 UTC to 2026-05-27 14:26:33 UTC.
Counts come from the repository’s dated source evidence. Index cutoff: 2026-11-02 08:16:59 UTC.
View mention counts and date ranges
| Period | Start (UTC) | End (exclusive, UTC) | Mentions | Submissions | Comments |
|---|---|---|---|---|---|
| May 26, 15:00 | 2026-05-26 15:00:00 UTC | 2026-05-26 16:00:00 UTC | 2 | 1 | 1 |
| May 26, 16:00 | 2026-05-26 16:00:00 UTC | 2026-05-26 17:00:00 UTC | 0 | 0 | 0 |
| May 26, 17:00 | 2026-05-26 17:00:00 UTC | 2026-05-26 18:00:00 UTC | 0 | 0 | 0 |
| May 26, 18:00 | 2026-05-26 18:00:00 UTC | 2026-05-26 19:00:00 UTC | 0 | 0 | 0 |
| May 26, 19:00 | 2026-05-26 19:00:00 UTC | 2026-05-26 20:00:00 UTC | 0 | 0 | 0 |
| May 26, 20:00 | 2026-05-26 20:00:00 UTC | 2026-05-26 21:00:00 UTC | 0 | 0 | 0 |
| May 26, 21:00 | 2026-05-26 21:00:00 UTC | 2026-05-26 22:00:00 UTC | 0 | 0 | 0 |
| May 26, 22:00 | 2026-05-26 22:00:00 UTC | 2026-05-26 23:00:00 UTC | 0 | 0 | 0 |
| May 26, 23:00 | 2026-05-26 23:00:00 UTC | 2026-05-27 00:00:00 UTC | 0 | 0 | 0 |
| May 27, 00:00 | 2026-05-27 00:00:00 UTC | 2026-05-27 01:00:00 UTC | 0 | 0 | 0 |
| May 27, 01:00 | 2026-05-27 01:00:00 UTC | 2026-05-27 02:00:00 UTC | 0 | 0 | 0 |
| May 27, 02:00 | 2026-05-27 02:00:00 UTC | 2026-05-27 03:00:00 UTC | 0 | 0 | 0 |
| May 27, 03:00 | 2026-05-27 03:00:00 UTC | 2026-05-27 04:00:00 UTC | 0 | 0 | 0 |
| May 27, 04:00 | 2026-05-27 04:00:00 UTC | 2026-05-27 05:00:00 UTC | 0 | 0 | 0 |
| May 27, 05:00 | 2026-05-27 05:00:00 UTC | 2026-05-27 06:00:00 UTC | 0 | 0 | 0 |
| May 27, 06:00 | 2026-05-27 06:00:00 UTC | 2026-05-27 07:00:00 UTC | 0 | 0 | 0 |
| May 27, 07:00 | 2026-05-27 07:00:00 UTC | 2026-05-27 08:00:00 UTC | 0 | 0 | 0 |
| May 27, 08:00 | 2026-05-27 08:00:00 UTC | 2026-05-27 09:00:00 UTC | 0 | 0 | 0 |
| May 27, 09:00 | 2026-05-27 09:00:00 UTC | 2026-05-27 10:00:00 UTC | 0 | 0 | 0 |
| May 27, 10:00 | 2026-05-27 10:00:00 UTC | 2026-05-27 11:00:00 UTC | 0 | 0 | 0 |
| May 27, 11:00 | 2026-05-27 11:00:00 UTC | 2026-05-27 12:00:00 UTC | 0 | 0 | 0 |
| May 27, 12:00 | 2026-05-27 12:00:00 UTC | 2026-05-27 13:00:00 UTC | 0 | 0 | 0 |
| May 27, 13:00 | 2026-05-27 13:00:00 UTC | 2026-05-27 14:00:00 UTC | 0 | 0 | 0 |
| May 27, 14:00 | 2026-05-27 14:00:00 UTC | 2026-05-27 15:00:00 UTC | 1 | 1 | 0 |
What Hacker News says
3 sourcesAVTR-1 A free, open-source, open weights real time avatar model
2 points · Open discussionHi HN — I'm Sergei, founder of Avaturn, our team built AVTR-1. It's an open-weights real-time avatar model built for real time interactions: feed it one reference photo and any audio, it generates a talking-head video frame by frame in real time, every pixel of the face every frame — no pre-recorded clip, no mouth swap. Sub-200ms end-to-end on one A100, runs well as well on 4060 or higher.
Code, : https://github.com/avaturn-live/avtr-1 and
Weights : https://huggingface.co/avaturn-live/avtr-1
Project page : https://avtr-1.avaturn.live
Live demo no sign ups: https://avaturn.live
Along side the model we release our streamer so you can theoretically drop in any other open-weight real-time video model, not just ours plug in any conversation backend, ship a low-latency experience without rebuilding orchestration.
If we gain enought traction
Read the rest
our next step is to covince other avatar providers that running a public leaderboard is useful. As of now, there isn't such for real-time avatars. We think transperancy is useful and can accelrate research for all. In the meantime we ran our own benchmark vs other real time models , with SOTA results on 5/6 subtests , I would be happy for you to take a look and suggest any improvments to our benchmarks. (Full results are on our project page)
The model is open weights under a community license — free for personal, research, and any commercial use under $10M ARR. I would most appreciate your feedback on it.