Zack Li

183 posts

Zack Li banner
Zack Li

Zack Li

@zacklearner

Co-founder & CTO @ Nexa AI (acquired by Qualcomm). Nexa SDK / GenieX · ex-Google, Amazon · Stanford. Building scalable AI from research to products.

Santa Clara, CA, US Katılım Ekim 2021
425 Takip Edilen239 Takipçiler
Zack Li retweetledi
Qualcomm
Qualcomm@Qualcomm·
Join us live to see what developers can build with NexaSDK on @Snapdragon—plus a live demo and open Q&A with the team. 🚀
English
0
8
10
3.1K
Zack Li
Zack Li@zacklearner·
What an incredible 2025! Thankful for the partnerships, trust, and collaboration that made it possible. Excited for what we’ll build together next 🚀
English
0
0
3
194
Zack Li
Zack Li@zacklearner·
gpt-oss-20b running on Hexagon NPU via Nexa SDK 🔥
English
0
0
0
83
AMD
AMD@AMD·
@nexa_ai Local AI, limitless potential. 🔥 Hyperlink is helping shape the next era of personal intelligence, accelerated by AMD Ryzen AI.
English
2
0
10
887
Ison Ward
Ison Ward@Isonwardster·
@nexa_ai Hyperlink feels like the missing layer between memory and action. It understands your local data, connects context instantly, and turns chaos into clarity. No cloud, no noise. Just you and your system working as one. Once you try it, going back feels impossible.😎
English
2
1
9
1.3K
Robin Delta
Robin Delta@heyrobinai·
@nexa_ai this is the AI that's making Google nervous
English
2
0
3
401
The SamurAI
The SamurAI@samuraipreneur·
@nexa_ai I was testing this yesterday, and it works actually really well. local AI is definitely the right move forward, everyone will have local ai on their devices en near future I think.
English
2
0
5
776
Qualcomm
Qualcomm@Qualcomm·
@nexa_ai Great to see @nexa_ai unlock a new era of personal, local intelligence. Can’t wait to see Hyperlink in action.
English
2
0
2
1.3K
Zack Li
Zack Li@zacklearner·
Following the launch of the Nexa Android SDK, we ran a 10-minute LLM stress test on the Samsung S25 Ultra with Qualcomm Hexagon NPU: ⚙️ CPU: throttled from ~37 t/s → ~19 t/s at 42 °C ⚙️ NPU (Qualcomm Hexagon): held steady at ~90 t/s and 36–38 °C — 2–4× faster under load 🔋 Both used ~6% battery, but the NPU delivered ~3.7× more tokens per battery Efficient, cool, and fully offline — that’s the power of NexaSDK on-device inference. 🔥📱 📘 Learn more: docs.nexa.ai/nexa-sdk-andro…
English
0
0
1
154
Zack Li
Zack Li@zacklearner·
Today, we’re launching Android Java & Kotlin support for NexaSDK (Beta) — bringing the full power of on-device AI to billions @Android phones powered by @Qualcomm @Snapdragon chipsets. This is a major leap forward for the world’s largest mobile developer community: ✅ Seamless NPU / GPU / CPU inference with a single SDK ✅ Run top-tier open models from @Google , @nvidia , and more ✅ 100% private, zero cloud latency, and fully offline
English
0
0
2
108