Guides

Local AI guides

Run local AI on your own hardware - practical, cost-driven guides.

Write a guide
Guide

Dessa: Block's speech model trained only on listening

Block (Jack Dorsey) unveiled Dessa, a speech-to-speech model trained only on human conversation. It hears tone, pauses and emotion, and talks while it listens.

Oct 23, 2026

Guide

Kimi K3.1 Pro: Moonshot's 3T-class multimodal model

Kimi K3.1 Pro: 2.8T total / 104B active, 1M context, multimodal. Terminal-Bench 4.0 65.7 near Opus 5.5; BenchCAD 92.7. Weights by Nov 4, cyber data removed.

Oct 23, 2026

Guide

Nemotron 3.5 Bolt E4B: the Minecraft NPC model

NVIDIA's Bolt-E4B-4GB (8B total, 4B effective) runs a Minecraft blacksmith NPC that haggles, smelts and crafts; also a robot on Jetson. OpenMDW, GGUF.

Oct 23, 2026

Guide

Base Labs mammoth-1 and elephant-1: open decision models

Base Labs shipped two open decision models: mammoth-1 (27B) and elephant-1 (9B), Qwen3.5 fine-tunes that output per-option probabilities in one pass. MIT.

Oct 23, 2026

Guide

Raon-1 mini: Krafton's 30B open model with thinking

Raon-1 mini is a 30B MoE with 4B active under Apache-2.0. Thinking channel and DFlash2 draft ship with it: NVFP4 for one RTX 5090, MLX for Apple, GGUF too.

Oct 23, 2026

Guide

Recursive Flint: 40B coding model for 48 GB laptops

Flint is a 40B MoE coding model with 2.5B active under Apache-2.0: 131 tok/s on an M4 Pro, 156 on an RTX Spark, built largely by an automated research system.

Oct 23, 2026

Guide

NVIDIA Flare: a Rust server for Spark-class local inference

NVIDIA's Flare serves SM120/SM121 GPUs from a Rust runtime built by PyTorch compile. Bolt 74B on one card, Qwen3.8 and DeepSeek V4.1 Flash recipes.

Oct 21, 2026

Guide

DeepInfer: DeepSeek open sourced its inference engine

DeepSeek's production inference engine DeepInfer is now open source under Apache-2.0, serving the V4.1 Flash model at scale on SM100 and Ascend 950.

Oct 20, 2026

Guide

The refusal tax: one business's $32,201.72 cloud answer

OpenAI's model started refusing his gun directory's normal work, so Nick Reese built 12 R9700 GPUs in-house for $32.2k and got 15x the throughput. The receipts.

Oct 19, 2026

Guide

DeepSeek banned out of US civilian IT, built-on tools too

A DHS FASCSA order removes DeepSeek and every third-party tool built on it from federal civilian IT. The MIT weights stay legal elsewhere.

Oct 19, 2026

Guide

LGTM: the AI satire video that approves itself 1,363 times

Degenerative Pixels' LGTM music video hit 1.2M views satirizing AI code review, and the joke is layered: Claude Opus 5.5 wrote the whole animation as code.

Oct 19, 2026

Guide

The fruit fly is AI's favorite research animal now

Six years of fly-brain mapping ended with the whole CNS in 2026, and builders put it in Doom, Mario 64 and a world model. What the demos actually show.

Oct 19, 2026