Baidu just dropped ERNIE-5.5-Flash-Base, and the weights are available. 280B MoE, only 14B active per token, pretrained from scratch on 33.5T tokens. It's a pure base model with no mid- or post-training, so labs can adapt it for domain-specific tasks, reasoning, and agents. What makes it exciting is how hard it pushes on efficiency: - Only 6 of 43 layers keep token-level access to the full context. The other 37 squeeze every 128 tokens into a single cached entry, which keeps the KV cache small. Context still goes up to 1M tokens for fine-tuning and inference. - Multi-Router MoE: several routers pick the experts together instead of just one, without increasing activated expert compute or communication costs. - ECHO: multi-token prediction runs through the model's own final layer, no separate MTP module needed. And the early numbers look strong. In Baidu's preliminary evals against DeepSeek-V4-Flash-Base, a model of almost the same size: - MMLU-Pro: 78.4 vs. 69.2 - HumanEval+: 82.3 vs. 69.5 - LiveCodeBench: 40.1 vs. 31.9 Weights are live on Hugging Face in FP8 and BF16: The model license excludes the EU, UK, and South Korea from its permitted territory.

Baidu killed the MTP module everyone copied from DeepSeek ERNIE 5.5 Flash just runs its own last layer a second time to guess the token after next. They call it ECHO. Trained on only 1 extra token and it still drafts several ahead for speculative decoding, w/ higher acceptance than an MTP module trained the same way With a normal MTP module the future token loss mostly trains the bolted on module. ECHO puts that loss on the real model so the backbone itself learns to look ahead, and Baidu says reasoning and code got clearly better. The leftover ECHO layer is like 10 times smaller than the MTP module tencent bolted on Hy3, a model the same size

Baidu is opening ERNIE-5.5-Flash-Base straight out of pre-training: an open-weight foundation for domain adaptation, instruction tuning, and RL research. 🤖 • 280B total parameters; 14B activated per token • Pre-trained on 33.5T tokens, with training context up to 128K • Up to 1M-token context for fine-tuning and inference • FP8 and BF16 weights 🛠️ A starting point for your own training: The checkpoint precedes all mid- and post-training, leaving developers to choose their own adaptation methods and objectives. 📊 Strong base-model results: Scores 78.4 on MMLU-Pro, 82.3 on HumanEval+, and 40.1 on LiveCodeBench, ahead of the three compared base models on these benchmarks under a unified evaluation pipeline. 🧠 Compressed attention for long inputs: Combines 37 HCA layers, each caching one entry per 128 tokens, with six MLA-DSA layers for full-context access through sparse attention. ⚙️ A built-in speculative draft: ECHO reuses the backbone's final layer to predict future tokens, replacing a separate MTP module. 📦 Deployment: Transformers, SGLang, and vLLM-ERNIE paths are available. The Base model uses the completions endpoint.

Baidu's MarathonDocBench evaluates the parsing of 100–300-page Chinese and English documents, including cross-page text, tables, and document structure. 🤖 📚 Hundred-page scale: 100 documents, approximately 18,000 pages, and 205,000 annotated layout elements. 🔗 Structure across pages: Human-refined annotations cover bounding boxes, heading hierarchy, reading order, and cross-page text and table continuations. 🧪 Document-level evaluation: Reuses MPDocBench-Parse's eight metrics to assess text, tables, formulas, reading order, figures, and heading structure. Overall is their mean. 📊 A demanding test: Listed general-purpose VLMs score below 34 Overall under the published whole-document protocol, which counts failed requests as empty predictions. Input and output limits also affect results. 📜 Apache-2.0 License.

Baidu Maps V22 is out🚀 Lane-Level Navigation 4.0 now covers residential compounds, campuses, parks, and parking lots — not just public roads. It brings continuous parking-space-to-parking-space navigation, plus 7 terrain types and 2 seasonal scenes. AI-powered image understanding also makes lanes and turning points clearer at complex intersections. SR Navigation 2.0 combines road data with real-time vehicle sensing, extending guidance into parking lots. It also adds AI-generated rendering while preserving real-world road information. The All-Domain Safety System supports 30+ safety alerts, detects major road anomalies within seconds, warns about vehicles approaching from behind, and reconstructs congestion scenes.

Baidu AI Cloud and World Agents inked a deal to launch the world's first 10,000-GPU-scale collective intelligence foundation model training initiative at the #WRC2026# on Aug. 19. The pair will build a collective AI Infra adaptable to diverse robot form factors, accelerating the commercial deployment of embodied intelligence across industrial, space, and other real-world scenarios.

Baidu CEO Robin Li said that the key metric for success in the AI era should be “daily active agents” rather than token consumption, underscoring his confidence in the transformative potential of AI agents in people’s everyday lives.