KyutaiCASA
Helium 1 VL 2B
This instruction-tuned reference model inserts image tokens into the text sequence. It is the comparison model for CASA’s cross-attention design.
Research release
The architecture determines how images enter the model
The reference keeps token insertion so researchers can compare it with the cross-attention models. It accepts an image and a text instruction and generates text.Helium1-VL-2B: Kyutai model card
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Research release |
| Access | Downloadable |
| Model identifiers | kyutai/Helium1-VL-2B |
| Weight license | CC BY-NC-SA 4.0.Helium1-VL-2B: Kyutai model card |
| Upstream components | Helium 1 and a Qwen 2.5 VL encoder. Gemma terms and the Qwen Research License apply.Helium1-VL-2B: Kyutai model cardhelium-1-2b: Kyutai model cardCASA-Qwen2_5-VL-3B: Kyutai model card |
Related models
- CASA Helium 1 VL 2B
Kyutai · CASA
A vision-language model that links Helium 1 with a Qwen 2.5 VL encoder through cross-attention.
Launch date not confirmedResearch releaseDownloadable