Zhenwei Gao 10/1/26 Zhenwei Gao 10/1/26

Disaggregated Inference from the Ground up

Authors: Isaac Tai and Zhenwei Gao

Read interactive blog here.

(400K X Impressions)

Zhenwei Gao 9/24/26 Zhenwei Gao 9/24/26

The Speed Advantage: Why Cyber Defense Needs Faster Inference

Cyber attackers move fast, so defenders have to move faster.

By removing inference as a bottleneck, Cerebras helps defenders act before attackers can advance.

(37K X impressions)

Zhenwei Gao 8/12/26 Zhenwei Gao 8/12/26

GPT-5.6-Sol-Ultrafast Financial Terminal Clone

We gave OpenAI GPT-5.6 Sol the same prompt on Ultrafast and Standard: build a financial terminal-style dashboard for analysts.

Ultrafast: 1 min 50 seconds
Standard: 12 min 20 seconds

Same result, nearly 7x faster.

Learn more about GPT-5.6 Sol Ultrafast in OpenAI’s blog post.

(145K X Impressions)

Zhenwei Gao 7/27/26 Zhenwei Gao 7/27/26

Getting the most out of GPT-5.6: Sol, Terra, and Luna

Your Codex subscription now comes with 3 main models, Sol, Terra, and Luna, each independently trained and served, with reasoning dials. Together, they let you choose a balance of speed, cost, and intelligence for each task.

(660K X Impressions)

Zhenwei Gao 7/15/26 Zhenwei Gao 7/15/26

Build ultra-fast LLM applications with Cerebras and DeepLearning.AI

Fast inference makes a new class of real-time LLM applications possible.

In our new short course, Fast LLM Inference with Cerebras, taught by Zhenwei Gao, Seb Duerr, and Sarah Chieng, you'll build them on the Cerebras Wafer-Scale Engine (WSE), where a model's weights sit on-chip and tokens come out several times faster than a typical GPU setup.

(243K X impressions)

Zhenwei Gao 7/8/26 Zhenwei Gao 7/8/26

First Look @ Gemma 4 on Cerebras

We spent a week building multimodal use cases with Gemma 4, running on Cerebras at ~2,300 tok/s. Here are some of the things we built and our learnings.

(28K X Impressions)