**[12/25/2024]** 🚀 We updated the [leaderboard](https://charxiv.github.io/#leaderboard) with the latest models: [o1](https://openai.com/o1/), [Qwen2-VL](https://github.com/QwenLM/Qwen2-VL), [Pixtral](https://mistral.ai/news/pixtral-12b/), [InternVL 2.5](https://internvl.github.io/blog/2024-12-05-InternVL-2.5/), [Llama 3.2 Vision](https://ai.meta.com/blog/llama-3-2-connect-2024-vision-edge-mobile-devices/), [NVLM](https://nvlm-project.github.io/), [Molmo](https://molmo.org/), [Llava OneVision](https://llava-vl.github.io/blog/2024-08-05-llava-onevision/), [Phi 3.5](https://techcommunity.microsoft.com/blog/azure-ai-services-blog/discover-the-new-multi-lingual-high-quality-phi-3-5-slms/4225280), and more!
**[10/24/2024]** 🚀 Check out [this paper](https://arxiv.org/abs/2410.18798) which includes a detailed study on 🪜 improving the model performance on CharXiv!
**[10/10/2024]** 🚀 CharXiv is accepted at [**NeurIPS 2024 Datasets & Benchmarks Track**](https://openreview.net/forum?id=cy8mq7QYae) and NeurIPS 2024 [Multimodal Algorithmic Reasoning Workshop](https://marworkshop.github.io/neurips24/) as a **spotlight** paper.
**[07/26/2024]** 🚀 Upcoming this week: we'll be releasing scores for [GPT-4o-mini](https://openai.com/index/gpt-4o-mini-advancing-cost-efficient-intelligence/) as well as the largest and most capable open-weight VLM in our benchmark: [InternVL2 LLaMA-3 76B](https://huggingface.co/OpenGVLab/InternVL2-Llama3-76B). Alongside scores, we find some [interesting patterns](https://x.com/zwcolin/status/1816948825036071196) in the trend of model improvement with respect to differnet chart understanding benchmarks on X.
**[07/24/2024]** 🚀 We released the [full evaluation pipeline](https://github.com/princeton-nlp/CharXiv)(i.e., v1.0).
