Zhipu CogVLM2 Achieves SOTA Performance in OCR Tasks
teortaxesTex · x · 2026-08-23
The author praises Zhipu's CogVLM, noting that the Llama 3-based V2 is SOTA tier in text recovery tasks. Features like 1344^2 resolution and a large encoder contribute to its performance. A related dataset has also been released.
Related event: Zhipu's CogVLM2 Hits SOTA on OCR Tasks(2 posts)→
More from Models
- Stefano Ermon teases 'largest and most capable model' landing next week, opens early access — StefanoErmon · 2026-10-16
- Inception teases its biggest model yet after Mercury 2 makes TIME's Best Inventions — StefanoErmon · 2026-10-16
- Diffusion LLM Mercury 2 named to TIME's Best Inventions of 2026 — StefanoErmon · 2026-10-16
- Gemini 4 Argon scores 97.5% on ARC-AGI-3, second lab past 95% milestone — fchollet · 2026-10-16
- New inference engine 'e' hits 10,500 tok/s on a single DGX Station — brandon_galang · 2026-10-16
- NVIDIA's Nemotron 3.5 Super VL hits CoreWeave: 12B-active multimodal MoE at $0.25/M input tokens — wandb · 2026-10-16