GMgmays·May 12, 2026Submission
Tuna-2: Pixel Embeddings Beat Vision Encoders for Understanding, Generation
1 points · Open discussionfacebookresearch
tuna-2
Python2026–2026
facebookresearch / tuna-2
tuna-2
Official implementation of Tuna-2: Pixel Embeddings Beat Vision Encoders for Unified Understanding and Generation
3Distinct people
3Mentions
0Explicit recommenders
763GitHub stars
What to know before exploring
What it is
Official implementation of Tuna-2: Pixel Embeddings Beat Vision Encoders for Unified Understanding and Generation
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 day3 dated mentions
Observed mentions: 2026-04-28 18:47:16 UTC to 2026-05-12 03:40:55 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 |
|---|---|---|---|---|---|
| Apr 28 | 2026-04-28 00:00:00 UTC | 2026-04-29 00:00:00 UTC | 1 | 0 | 1 |
| Apr 29 | 2026-04-29 00:00:00 UTC | 2026-04-30 00:00:00 UTC | 0 | 0 | 0 |
| Apr 30 | 2026-04-30 00:00:00 UTC | 2026-05-01 00:00:00 UTC | 0 | 0 | 0 |
| May 1 | 2026-05-01 00:00:00 UTC | 2026-05-02 00:00:00 UTC | 0 | 0 | 0 |
| May 2 | 2026-05-02 00:00:00 UTC | 2026-05-03 00:00:00 UTC | 1 | 1 | 0 |
| May 3 | 2026-05-03 00:00:00 UTC | 2026-05-04 00:00:00 UTC | 0 | 0 | 0 |
| May 4 | 2026-05-04 00:00:00 UTC | 2026-05-05 00:00:00 UTC | 0 | 0 | 0 |
| May 5 | 2026-05-05 00:00:00 UTC | 2026-05-06 00:00:00 UTC | 0 | 0 | 0 |
| May 6 | 2026-05-06 00:00:00 UTC | 2026-05-07 00:00:00 UTC | 0 | 0 | 0 |
| May 7 | 2026-05-07 00:00:00 UTC | 2026-05-08 00:00:00 UTC | 0 | 0 | 0 |
| May 8 | 2026-05-08 00:00:00 UTC | 2026-05-09 00:00:00 UTC | 0 | 0 | 0 |
| May 9 | 2026-05-09 00:00:00 UTC | 2026-05-10 00:00:00 UTC | 0 | 0 | 0 |
| May 10 | 2026-05-10 00:00:00 UTC | 2026-05-11 00:00:00 UTC | 0 | 0 | 0 |
| May 11 | 2026-05-11 00:00:00 UTC | 2026-05-12 00:00:00 UTC | 0 | 0 | 0 |
| May 12 | 2026-05-12 00:00:00 UTC | 2026-05-13 00:00:00 UTC | 1 | 1 | 0 |
What Hacker News says
3 sourcesPixel Embeddings Beat Vision Encoders for Unified Understanding and Generation
13 points · Open discussionInteresting note from the repo [1]:
> A Note on Model Release
> Due to organizational policy constraints, we are unable to release the full production-trained model weights. To support the research community, we plan to release a foundation checkpoint with a small number of layers removed from both the LLM backbone and the diffusion head (flow head). The remaining layers and all other components (vision encoder, projections, embeddings, etc.) are fully preserved. With a short fine-tuning pass on your own data, the removed layers can be quickly re-learned and the model restored to full quality.
[1] https://github.com/facebookresearch/tuna-2#a-note-on-model-r...
In Tuna-2: VAE-less image model from Meta