- [ ] Support training using 3090 GPUs (24Gb video memory)
- [ ] Improve Chinese language proficiency
- [ ] Improve Chinese language proficiency
- [ ] TextMonkey with different LLMs
- [ ] TextMonkey with different LLMs
# Model Zoo
# Model Zoo
TextMonkey was trained using 8 A800 GPUs on a dataset of just 40k image-text pairs, requiring approximately 1 day and 6 hours of training time. It is capable of running inference on a 3090 GPU.