|
--- |
|
license: apache-2.0 |
|
--- |
|
|
|
# Haplo Model Card |
|
|
|
This work presents a simple yet efficient method to construct a baseline for the native and end-to-end large multi-modal model in a single transformer. The proposed model demonstrates superior performance compared to other LMMs using one transformer and significantly narrows the performance gap with compositional LMMs. |
|
|
|
**Model date:** |
|
Haplo models were trained in September 2024. |
|
|
|
**Paper or resources for more information:** https://haplo-vl.github.io/ |
|
|
|
**Performace** |
|
|
|
| Model | SEEDB | POPE | RWQA | MMB | MMStar | VQAv2 | GQA | SQA | |
|
|------------|-------|------|------|-----|--------|-------|-----|------| |
|
| Haplo-8B-672 | 75.1 | 88.6 | 61.4 | 73.6| 57.2 | 81.0 | 65.5| 95.3 | |
|
| Haplo-8B-MI-672 | 75.5 | 88.2 | 62.0 | 75.0| 57.6 | 80.7 | 65.0| 94.4 | |
|
|
|
|
|
## Intended use |
|
**Primary intended uses:** |
|
The primary use of Haplo is research on large multimodal models and chatbots. |
|
|
|
**Primary intended users:** |
|
The primary intended users of the model are researchers and hobbyists in computer vision, natural language processing, machine learning, and artificial intelligence. |
|
|
|
|
