List your company
ResearchFormerly WizWand

Text-Centric Vision-Language Understanding on OCR Bench

517AccuracyInternVL
47.96169.73291.5413.27Jul 23, 2024Oct 13, 2024Jan 4, 2025Mar 28, 2025Jun 19, 2025Sep 10, 2025Dec 2, 2025
Updated 8mo ago

Evaluation results

MethodLinks
InternVLModel Generation Capability=Text only2024.07517
MonkeyModel Generation Capability=Text only2024.07514
InternLM-XComposer2Model Generation Capability=Text only2024.07511
TextHarmony-ChatModel Generation Capability=Text and image, Slide-LoRA=true2024.07448
TextHarmonyModel Generation Capability=Text and image, Slide-LoRA=true2024.07440
TextHarmony*Model Generation Capability=Text and image, Slide-LoRA=false2024.07397
mPLUG-Owl2Model Generation Capability=Text only2024.07366
SEED-LLaMA-14BModel Generation Capability=Text and image2024.07357
LLaVARModel Generation Capability=Text only2024.07346
LLaVA1.5-7BModel Generation Capability=Text only2024.07297
MM-InterleavedModel Generation Capability=Text and image2024.07197
Qwen2.5-VL-3B-InstructParams=3.75B2025.1282.7
Qwen2-VL-2BParams=2.2B2025.1280.9
InternVL2-2BParams=2.2B2025.1278.4
InternVL2-1BParams=0.94B2025.1275.7
MobileNet-QwenParams=1.84B2025.1273.5
InternViT-QwenParams=1.86B2025.1272.5
AutoNeuralParams=1.47B2025.1271.4
MiniGPT5Model Generation Capability=Text and image2024.0768
InternViT-LiquidParams=1.49B2025.1266