15:56
MiniCPM-V 4.6: The Agent Vision Model
Sam Witteveen examines MiniCPM-V 4.6, a 1.3 billion parameter vision-language model released by OpenBMB—a joint initiative between AI...
Sort by: Newest Items
There are 89 items in this page
08:06
MTP vs DFlash — Speculative Decoding Explained Simply
This video by Fahd Mirza offers a clear, structured comparison of two speculative decoding techniques — Multi-Token Prediction (MTP)...
43:11
Local Hermes & Openclaw on Beelink in 43 mins
Keith AI delivers a detailed, framework-driven evaluation of running Hermes and OpenClaw locally on a Beelink S10 Max mini PC — the d...
08:28
Qwen3-8B at 74 tok/s with RedHat DFlash Speculator on vLLM Locally
Fahd Mirza walks through running Red Hat's DFlash speculative decoding implementation on Qwen3-8B using vLLM, achieving 74 tokens per...
11:00
NVIDIA Nemotron Elastic: 3-in-1 Elastic LLM Like Russian Dolls in One File
NVIDIA's Nemotron Elastic model family packs three reasoning models — 30B, 23B, and 12B parameters — into a single checkpoint file us...
13:37
Zaya1 8B – Intelligence Efficiency by Zyphra – Run Locally
Zyphra, a San Francisco AI lab known for earlier releases like Zonos and ZR1, has returned with Zaya 1 (Zia) 8B — an open-source mixt...
08:43
DFlash Drafter for Gemma 4 26B – Official Speculative Decoding is Here: Run Locally
ZLab, the UC San Diego research team that invented DFlash speculative decoding, has released the first official drafter model paired...
08:41
Gemma 4 31B at 196 tok/s with RedHat DFlash Speculator Locally
This hands-on tutorial from the Fahd Mirza channel demonstrates running Google's Gemma 4 31B model locally at 196 tokens per second u...
32:36
RTX 5090, Mac Studio, or DGX Spark? I tried all three.
Nate B Jones tests the RTX 5090, Apple Mac Studio, and NVIDIA DGX Spark as personal AI computing platforms, but the video is as much...
09:01
Running a 27B model at 130 tokens sec on a single GPU Locally with Luce DFlash
LlamaDeFlash is a custom inference engine built from scratch in C++ and CUDA — no vLLM, no llama.cpp, no Python in the critical path...