# Blog

## Agentic AI 
**The Next Chapter: Clarifai Compute Orchestration and Reasoning Engine Joins Nebius**  
Today, we are thrilled to announce a monumental milestone in Clarifai’s journey: we have entered into an agreement for Nebius (NASDAQ: NBIS), the AI cloud company, to license our core engineering a...  
**Date:** May 19, 2026  
**Author:** Matthew Zeiler, Ph.D. Founder and CEO, Clarifai

## Documentation 
**NVIDIA Nemotron 3 Nano Omni on Clarifai Reasoning Engine: Zero Day Support at 400 Tokens Per Second**  
We’re excited to announce day-0 support for NVIDIA Nemotron 3 Nano Omni on Clarifai. Available now on [Clarifai Reasoning Engine](/content/nvidia/chat-completion/models/nemotron-nano-v3...  
**Date:** April 28, 2026  
**Author:** Clarifai

## Documentation 
**Clarifai 12.3: Introducing KV Cache-Aware Routing**  
*This blog post focuses on new features and improvements. For a comprehensive list, including bug fixes, please see the release notes.*  
**Date:** April 9, 2026  
**Author:** Clarifai

## Clarifai API 
**run gemma 4 locally deploy frontier ai on your hardware with public api access**  
When you want to run frontier models locally, you hit the same constraints repeatedly.  
**Date:** April 7, 2026  
**Author:** Clarifai

## Documentation 
**what is kimi k2.5 architecture benchmarks ai infra guide**  
Open‑weight models are rapidly narrowing the gap with closed commercial systems. **As of early 2026**, Moonshot AI’s **Kimi K2.5** is the flagship of this trend: a **one‑trillion parameter Mixture‑...  
**Date:** March 18, 2026  
**Author:** Clarifai

## Documentation 
**ilama.cpp**  
Local large‑language‑model (LLM/index.html) inference has become one of the most exciting frontiers in AI. As of 2026, powerful consumer GPUs such as NVIDIA’s RTX 5090 and Apple’s M4 Ultra enable **state‑of‑t...  
**Date:** March 17, 2026  
**Author:** Clarifai

## Documentation 
**Clarifai Reasoning Engine Achieves 414 Tokens Per Second on Kimi K2.5**  
Using custom CUDA kernels and speculative decoding optimized for reasoning workloads, we achieved 414 tokens per second throughput on Kimi K2.5 running on Nvidia B200 GPUs, making us one of the fir...  
**Date:** March 16, 2026  
**Author:** Clarifai

## Documentation 
**flash attention 2**  
The transformer revolution is now deep into its long‑context era. **Models like GPT‑4 (32 k tokens), MosaicML’s MPT (65 k), and Claude (100 k)** can process entire chapters or codebases. Yet as con...  
**Date:** March 16, 2026  
**Author:** Clarifai

## Documentation 
**Clarifai 12.2: Three-Command CLI Workflow for Model Deployment**  
*This blog post focuses on new features and improvements. For a comprehensive list, including bug fixes, please see the release notes.*  
**Date:** March 12, 2026  
**Author:** Clarifai

## Documentation 
**clarifai vs other inference providers**  
The AI landscape of **2026** is defined less by model training and more by how effectively we serve those models. The industry has learned that inference—the act of deploying a pre‑trained model—is...  
**Date:** March 10, 2026  
**Author:** Clarifai

## Documentation 
**what is lpu**  
The AI hardware landscape is shifting rapidly. Five years ago, GPUs dominated every conversation about AI acceleration. Today, agentic AI, real‑time chatbots and massively scaled reasoning systems ...  
**Date:** March 10, 2026  
**Author:** Clarifai