Cisco Model Provenance Kit: A DNA Test for Transformer Weights
Cisco open-sourced a Python toolkit that fingerprints model weights to determine lineage. Installation, CLI walkthrough, and real comparison between models.
Archive
All articles
Evergreen writing on AI tools, workflows, tutorials, reviews, comparisons, and technical explainers. If you want the fastest path in, start with the homepage's Start Here section.
GLM-5.2 vs Kimi K2.7-Code: Which Open-Weight Coding Model to Use?
Head-to-head comparison of GLM-5.2 and Kimi K2.7-Code on SWE-bench, pricing, context window, and real-world testing — which open-weight coding model fits your workflow?
OpenRouter Fusion: Match Fable 5 on Deep Research at Half the Cost
OpenRouter Fusion synthesizes multiple AI models into one answer. A budget panel scored within 1% of Fable 5 at half the cost — benchmarks, API, and decision guide.
AI Supply Chain Security: Model Provenance and Provider Risk
The Fable 5 ban and LiteLLM backdoor proved AI supply chain risk is real. How to verify model provenance, audit dependencies, and monitor providers.
Multi-Provider AI Gateways: Never Let a Model Ban Break Your Stack
Practical guide to OpenRouter and LiteLLM fallback routing — survive Fable 5-style shutdowns, rate limits, and provider outages without changing code.
GLM-5.2: 1M Context and IndexShare Sparse Attention for Coding
Technical guide to GLM-5.2 — Z.ai's open-weight 744B MoE model with 1M token context, IndexShare sparse attention, and local execution via llama.cpp.
Kimi K2.7-Code: Open-Weights 1T MoE Coding Model
Hands-on analysis of Moonshot AI's K2.7-Code — 384-expert MoE, $0.95/M tokens, beats Opus 4.8 on MCP tool calling, with API walkthrough and deployment guide.
Introducing Open-TechStack: Practical AI Analysis for Builders
Our launch editorial explaining what Open-TechStack is all about: practical, benchmarked guidance on AI models, open-source code, and workflows.