InternLM-XComposer Models by Intern-AI
Last refreshed 2026-05-19. Next refresh: weekly.
Details
About
The InternLM-XComposer family features a suite of advanced large vision-language models (LVLMs) tailored for sophisticated text-image understanding and composition. These models showcase exceptional capabilities in multimodal tasks, performing on par with GPT-4V while utilizing a more compact 7B parameter language model backend. Prominent functionalities include the interpretation of ultra-high resolution images, detailed video analysis, and handling complex multi-turn, multi-image dialogues. Additionally, they can convert text or image instructions into web pages and produce high-quality text-image articles. As open-source models, they are readily accessible for further exploration and innovation in the research community. The latest version, InternLM-XComposer-2.5, offers enhanced performance over its predecessors, particularly in managing longer context scenarios 24.
Decision facts
- Best fit
- coding
- Capability starting point
- InternLM XComposer 7B with 4k context
- Lowest tracked input
- Not tracked
- Closest related family
- InternVL
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
Use when the workload needs 4k context and 7B parameters.
Use when the workload needs 4k context and 1.7B parameters.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| InternLM XComposer 7B | Use when the workload needs 4k context and 7B parameters. | 2023-10 | 4k context7B parameters | Current |
| InternLM XComposer 1.7B | Use when the workload needs 4k context and 1.7B parameters. | 2023-10 | 4k context1.7B parameters | Current |
Release Timeline
1 release groupSpecifications(2 models)
| Model | Released | Context | Parameters |
|---|---|---|---|
| InternLM XComposer 7B | 2023-10 | 4k | 7B |
| InternLM XComposer 1.7B | 2023-10 | 4k | 1.7B |
