NaiveAI

100× intelligence for the pioneers.

[NaiveAI @naiveailab]

@teortaxesTex Yes, we credited DeepSeek in our acknowledgments. This is our first open-source release, and we’ll keep working to contribute to the community. BTW, we’re not a Singaporean team.

[NaiveAI @naiveailab]

@_shiqiansu Models that train models is the future

[NaiveAI @naiveailab]

@EmperorMK_01 Yes, it's MiMo-V2.5 as the base — credited in our tech blog. This is what open source makes possible.

[NaiveAI @naiveailab]

Introducing Naive-N0.5-Flash: Building Frontier AI with AI

🔹 309B MoE, 15.5B active: top-tier in coding, leading in AI R&D.
🔹 Native 1M context, no full-attention layers (hybrid SWA + DSA).
🔹 Inference runtime built by AI: up to 2,000 tok/s in Ultrafast mode.

Weights are open today under MIT license.

🔗 Tech blog: https://t.co/P1e8pBeJJM
🤗 Hugging Face: https://t.co/v2YvKfuguc
💻 GitHub: https://t.co/QjJ7cjFlkL
🌐 https://t.co/sIc7nZJH1L

[NaiveAI @naiveailab]

@canberkys Thanks for it! Fixed now

[NaiveAI @naiveailab]

@paru_saha everyone likes the speed!

[NaiveAI @naiveailab]

@Xianbao_QIAN Thanks for the support! This is just our first model — more to come.