Qwen-SEA-LION-v4-8B-VL
Qwen-SEA-LION-v4-8B-VL is the eight-billion-parameter member of the v4 Qwen vision pair. It builds on Qwen3-VL-8B-Instruct and adds about nine million instruction-text pairs for regional language adaptation.
How this release works
It retains the Qwen3-VL tokenizer and tool-use capability. Its language set covers English and seven regional languages. The model card reports both regional language assessments and checks of retained vision-language capability.Qwen-SEA-LION-v4-8B-VL official model card
How to assess it for your work
Practical guidance. Compare the 8B and 4B releases on the same image sequence and questions. For document work, record which page supports each answer. A long context allowance can make a large input possible, but it does not establish reliable retrieval of every detail across that input.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | aisingapore/Qwen-SEA-LION-v4-8B-VL |
| Base model | Qwen/Qwen3-VL-8B-InstructQwen-SEA-LION-v4-8B-VL official model card |
| Context stated by the card | 256K tokensQwen-SEA-LION-v4-8B-VL official model card |
| License declared by the model card | MITQwen-SEA-LION-v4-8B-VL official model card |
| Input and output | Image and text inputs; text outputQwen-SEA-LION-v4-8B-VL official model card |
Related models
- Gemma-SEA-LION-v4-27B-VL
AI Singapore · SEA-LION v4
A 27B Gemma-based vision-language model trained on regional image-text instructions with SEACrowd.
Launch date not confirmedDownloadable - Qwen-SEA-LION-v4-4B-VL
AI Singapore · SEA-LION v4
A 4B Qwen3-VL adaptation for regional image and text tasks with tool-use support.
Launch date not confirmedDownloadable - Qwen-SEA-LION-v4.5-27B-IT
AI Singapore · SEA-LION v4.5
A 27B multimodal model based on Qwen3.6, distilled for Southeast Asian language tasks.
Launch date not confirmedDownloadable