Baidu AI

Baidu's natively omni-modal flagship unveiled at Baidu World 2025 — 2.4T-param MoE processing text, images, audio, and video in one model.

Baidu's upgraded lightweight MoE model refined to boost reasoning depth and quality. Top-tier performance in logical puzzles, math, science, coding, text generation, and expert-level academic benchmarks.

Baidu's text-based MoE model with 21B total parameters and 3B activated per token. Exceptional multimodal understanding through heterogeneous MoE structures and modality-isolated routing.

Baidu's multimodal MoE chat model with 28B total parameters and 3B activated per token. Exceptional text and vision understanding through heterogeneous MoE with modality-isolated routing.

Baidu's multimodal MoE model (424B total / 47B active), open-sourced under Apache 2.0.

Baidu deep-thinking model matching DeepSeek R1 performance at half the price.