11:11
MiniCPM5-1B: New 1B King for Local AI – Full Demo
Fahd Mirza walks through a complete local installation and live evaluation of MiniCPM 5 in its 1 billion parameter variant, released...
Sort by: Newest Items
There are 95 items in this page
10:10
Intern-S2-Preview FP8: 35B Scientific Multimodal Model Running Locally
InternLM's latest release, Intern-S2-Preview, is a 35-billion-parameter scientific multimodal model that takes a different approach t...
10:48
LM Studio Just Got MTP — Qwen3.6-27B Runs 63% Faster with One Toggle
Fahd Mirza demonstrates how to enable Multi-Token Prediction (MTP) speculative decoding in LM Studio's new beta release (version 0.4....
17:48
Scenema Audio: AI Voice That Actually Performs – Rage, Grief, Joy in One Generation Locally
Fahd Mirza installs and tests Cinema Audio, a new expressive text-to-speech model extracted from LTX Video 2.3's 22-billion-parameter...
10:54
Talkie: I Ran a 1930 AI Model Locally and Talked to People from the Past
Fahd Mirza explores Talkie, a 13 billion parameter language model built by Alec Radford — the researcher behind GPT-2 — that was trai...
08:41
Luce DFlash Meets OpenClaw – Local AI Agents at 2x Speed with Qwen3.6-27B
Fahd Mirza walks through a complete, reproducible integration of DFlash — a speculative decoding inference engine — with OpenClaw, an...
09:22
DramaBox – Run Most Expressive TTS with Voice Cloning Locally
Fahd Mirza takes a hands-on look at DramaBox, a newly released expressive text-to-speech model that can be run locally on consumer-gr...
09:45
TurboQuant + DFlash: Supercharge Local LLM Speed
Fahd Mirza demonstrates the practical integration of two recently released local inference tools: Google Research's TurboCore KV cach...
11:12
Qwen3.6 27B Gets 20% Faster with MTP and llama.cpp Locally
Fahd Mirza demonstrates how to enable multi-token prediction (MTP) on Qwen3.6 27B using ik_llama.cpp — a community fork of the popula...
09:15
ZAYA1-VL-8B: Efficient Open Visual Intelligence – Run Locally
Fahd Mirza puts ZAYA1-VL-8B — the new vision-language model from Zeffa — through its paces on an NVIDIA RTX 6000 with 48GB of VRAM, s...