AI article
Yandex open-sourced an 80B model trained from scratch: what's inside and where it wins
Community description: AliceAI-Foundation-80B-A3B: a from-scratch MoE model under Apache 2.0, benchmarked against Qwen3.5, Nemotron-3 and DeepSeek-V4 Flash.
Dev.to | Sep 21, 2026 | Kirill Lukyanov
Automated excerpt
It's a base model — not instruction-tuned, not a chatbot. Only every fourth layer is classic (gated) attention. Yandex compared base versions against open models in the same weight class and above.
Selected automatically from source text; not independently written or fact-checked. Read the original for full context.