Another open-source model has arrived, and it’s even better than DeepSeek-V3.
The Allen Institute for AI just introduced Tülu 3 (405B) 🐫, a post-training model that is a fine-tune of Llama 3.1 405B, which outperforms DeepSeek V3.
Dario (Anthropic CEO) thinks we’ll see a billion dollar company with 1 employee in 2026.
His advice:
“Be ambitious. Even if it doesn’t work, another model will likely come out in 1-3 months that’ll make it work.”