Technology Innovation Institute

company

https://www.tii.ae/

AI & ML interests

Large language models

Recent Activity

SofianChay updated a collection 8 days ago

SigLino: Vision Foundation Models (SigLIP2 + DINOv3)

lkhphuc new activity 8 days ago

tiiuae/Falcon-Perception:[Solved] Unable to run test example (Torch OOM)

FalconPerception new activity 9 days ago

tiiuae/Falcon-OCR:Discussion: Highlighting PP-DocLayoutV3's contribution in your project

View all activity

Papers

Falcon Perception

Learnable Multipliers: Freeing the Scale of Language Model Matrix Layers

View all Papers

Articles

Falcon Perception

Alyah ⭐️: Toward Robust Evaluation of Emirati Dialect Capabilities in Arabic LLMs

Introducing Falcon H1R 7B

Introducing Falcon-H1-Arabic: Pushing the Boundaries of Arabic Language AI with Hybrid Architecture

📚 3LM: A Benchmark for Arabic LLMs in STEM and Code

Announcing NeurIPS 2025 E2LM Competition: Early Training Evaluation of Language Models

Falcon-H1: A Family of Hybrid-Head Language Models Redefining Efficiency and Performance

Falcon-Arabic: A Breakthrough in Arabic Language Models

Falcon-Edge: A series of powerful, universal, fine-tunable 1.58bit language models.

View all articles

updated a collection 8 days ago

SigLino: Vision Foundation Models (SigLIP2 + DINOv3)

Vision encoders distilled from DINOv3 and SigLIP2 (MoE & Dense). CVPR 2026. • 6 items • Updated 8 days ago • 17

in tiiuae/Falcon-Perception 8 days ago

[Solved] Unable to run test example (Torch OOM)

#1 opened 14 days ago by

FalconPerception

in tiiuae/Falcon-OCR 9 days ago

Discussion: Highlighting PP-DocLayoutV3's contribution in your project

#11 opened 10 days ago by

Finetuning method

#10 opened 10 days ago by

in tiiuae/siglino-0.6B 9 days ago

Update to image feature extraction

#2 opened 9 days ago by

updated 4 models 9 days ago

tiiuae/siglino-moe-0.15-0.6B

Image Feature Extraction • 0.5B • Updated 9 days ago • 136 • 7

tiiuae/siglino-moe-0.3-0.6B

Image Feature Extraction • 0.5B • Updated 9 days ago • 157 • 7

tiiuae/siglino-30M

Image Feature Extraction • 47.1M • Updated 9 days ago • 223 • 6

tiiuae/siglino-70M

Image Feature Extraction • 69.5M • Updated 9 days ago • 167 • 6

in tiiuae/siglino-0.6B 9 days ago

Add library name, update to "image-feature-extraction"

#1 opened 9 days ago by

in tiiuae/Falcon-OCR 10 days ago

Discussion: Highlighting PP-DocLayoutV3's contribution in your project

#11 opened 10 days ago by

authored a paper 11 days ago

Evaluating Arabic Large Language Models: A Survey of Benchmarks, Methods, and Gaps

Paper • 2510.13430 • Published Oct 15, 2025 • 1

authored 2 papers 11 days ago

Falcon-H1: A Family of Hybrid-Head Language Models Redefining Efficiency and Performance

Paper • 2507.22448 • Published Jul 30, 2025 • 71

3LM: Bridging Arabic, STEM, and Code through Benchmarking

Paper • 2507.15850 • Published Jul 21, 2025 • 6

authored 3 papers 11 days ago

3LM: Bridging Arabic, STEM, and Code through Benchmarking

Paper • 2507.15850 • Published Jul 21, 2025 • 6

NeurIPS 2025 E2LM Competition : Early Training Evaluation of Language Models

Paper • 2506.07731 • Published Jun 9, 2025 • 2

Falcon2-11B Technical Report

Paper • 2407.14885 • Published Jul 20, 2024 • 1

authored a paper 11 days ago

Evaluating Arabic Large Language Models: A Survey of Benchmarks, Methods, and Gaps

Paper • 2510.13430 • Published Oct 15, 2025 • 1

authored a paper 11 days ago

Are Arabic Benchmarks Reliable? QIMMA's Quality-First Approach to LLM Evaluation

Paper • 2604.03395 • Published 15 days ago • 2

authored a paper 11 days ago

Are Arabic Benchmarks Reliable? QIMMA's Quality-First Approach to LLM Evaluation

Paper • 2604.03395 • Published 15 days ago • 2