Atlas

Models

← All models

InternVL3-1B

OpenGVLab · 2025-04-11 · 938M parameters

InternVL3-1B is OpenGVLab's 938,193,024-parameter vision-language checkpoint, pairing an InternViT-300M vision encoder with a Qwen2.5-0.5B language backbone through a two-layer MLP projector. Released after native multimodal pre-training, supervised fine-tuning, and mixed preference optimization, it accepts text, image, and sampled-video inputs and generates text. Its repository declares Apache-2.0, while the surrounding InternVL project code is MIT.

Benchmark scores

BenchmarkScore
AI2D (OpenVLM)69.7
Capability103.9
CharXiv Reasoning21.0
CharXiv-D47.1
HallusionBench (OpenVLM)37.2
MATH-Vision18.8
MathVista (OpenVLM)46.9
MM-Vet (OpenVLM)58.7
MMMU Validation (OpenVLM)43.2
MMStar (OpenVLM)52.3
Loading Atlas data…