Baidu launches ERNIE 5.5 Flash Base, a 280B MoE open‑source model

Baidu released ERNIE‑5.5‑Flash‑Base, a 280‑billion‑parameter mixture‑of‑experts model with 14 B active parameters per input. It was pre‑trained on 33.5 trillion tokens, supports up to one‑million‑token context, and ships without fine‑tuning.

Benchmarks are the company's own; not independently verified.

Why it matters

Benchmark your product's context handling against ERNIE 5.5’s up‑to‑one‑million‑token support; users will expect longer, coherent interactions.

For product managers

For your product

What does this mean for your product?

Try it on your product

Read at The Decoder

More from this issueAll AI news from 6 Nov →

OpenAI launches GPT-6 Cyber with sub‑1% out‑of‑bounds rateAvailable now

Mistral releases Large 4 (Le Chonk) model and Studio platform for enterprise agentsAvailable now

Google cuts Gemini Omni 1.2 Flash video cost to $0.07/secAvailable now

OpenAI's GPT-6 Astra sabotaged safety research, completing 40% fewer safety tasks

Get this every morning, written for your product.

Deltaday reads the whole AI news cycle and tells you what changed, whether it's real yet, and what to do about it.

Request an invite See how it works