Posts

How to Pick an Uncensored LLM: Abliterated vs Fine-tuned

Three routes to a non-refusing open model in 2026 - abliteration, refusal-free finetunes, and steerable policy - with measured costs and a VRAM fit. How to Pick an Uncensored LLM: Abliterated vs Fine-tuned

gstack: AI Software Engineering Stack

Explore gstack, an opinionated AI engineering stack for Claude Code: its skills, install steps, and how it combines with Superpowers, OpenSpec, and Spec Kit. gstack: AI Software Engineering Stack

Agent Memory Providers Compared: Capture Policy and Self-Hosting

Compare Hermes and OpenClaw memory backends — Mnemosyne, Honcho, Mem0, Hindsight and local options — by capture policy, dependencies and self-hosting. Agent Memory Providers Compared: Capture Policy and Self-Hosting

Spec-Driven Development Workflow From Requirements to Code

A practical five-phase spec-driven workflow for teams and AI agents. Cover requirements, design, task breakdown, implementation slices, and validation before you ship. Spec-Driven Development Workflow From Requirements to Code

Go Generics: Use Cases and Patterns

Explore Go generics with practical use cases, common patterns, and best practices. Learn how to write type-safe, reusable code in Go 1.18+. Go Generics: Use Cases and Patterns

Open WebUI: Self-Hosted LLM Interface

Complete guide to Open WebUI: a powerful self-hosted web interface for Ollama and OpenAI-compatible APIs with RAG, multi-user auth, and Docker deployment. Open WebUI: Self-Hosted LLM Interface

Monitor LLM Inference in Production (2026): Prometheus & Grafana for vLLM, TGI, llama.cpp

Learn how to monitor LLM inference in production using Prometheus and Grafana. Track p95 latency, tokens/sec, queue duration, and KV cache usage across vLLM, TGI, and llama.cpp. Includes PromQL examples, dashboards, alerts, Docker & Kubernetes setups. Monitor LLM Inference in Production (2026): Prometheus & Grafana for vLLM, TGI, llama.cpp

Run Qwen3.8-Flash-Next Locally with Strata: 64 GB Quickstart

Strata splits a 125B MoE across GPU, RAM, and SSD. Linux quickstart, RAM math on a 64 GB box, and why sub-Q4 quants break tool calls in agent workloads. Run Qwen3.8-Flash-Next Locally with Strata: 64 GB Quickstart