Chameleon: Mixed-Modal Early-Fusion Foundation Models · HackerPapers discussions

Chameleon: Mixed-Modal Early-Fusion Foundation Models — mentioned by spott

Wed, 03 Jun 2026 17:28:34 GMT

This is just early fusion basically. FAIR did this 2 years ago now: https://arxiv.org/abs/2405.09818 I've been waiting for something like this to be released since then. The annoying thing is that chameleon was multi-modal out based on the same principles, but this model is just inputs... (I'm curious how they did pre-training without having multi-modal outputs as well. I wonder if they just chopped them off rather than support image output).

Paper and discussion history. A mention is not an endorsement.

Chameleon: Mixed-Modal Early-Fusion Foundation Models — mentioned by joelburget

Thu, 11 Jul 2024 01:04:18 GMT

Vision Transformers do a shocking amount of compression in the tokenizer. In the [Chameleon paper](https://arxiv.org/pdf/2405.09818) they say the tokenizer "encodes a 512 × 512 image into 1024 discrete tokens from a codebook of size 8192". That's 256 pixels per token (512 * 512 / 1024). If we assume that a pixel is 24 bits (3x 8 bit channels), this implies that they've compressed 256 * 24 = 6144 bits into 13 = (log2(8192)). [An Image is Worth 32 Tokens for Reconstruction and Generation](https://yucornetto.github.io/projects/titok.html) pushes this even further. If these models work similarly…

Paper and discussion history. A mention is not an endorsement.

Chameleon: Mixed-Modal Early-Fusion Foundation Models — HN submission

Tue, 21 May 2024 01:37:53 GMT

Chameleon: Meta’s New Multi-Modal LLM

304 points; 40 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Chameleon: Mixed-Modal Early-Fusion Foundation Models — HN submission

Mon, 20 May 2024 02:24:02 GMT

Chameleon: Mixed-Modal Early-Fusion Foundation Models

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Chameleon: Mixed-Modal Early-Fusion Foundation Models — HN submission

Fri, 17 May 2024 10:46:17 GMT

Chameleon: Mixed-Modal Early-Fusion Foundation Models

6 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Chameleon: Mixed-Modal Early-Fusion Foundation Models — HN submission

Fri, 17 May 2024 01:11:42 GMT

Chameleon: Mixed-Modal Early-Fusion Foundation Models

4 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.