Guides
Practical guides to running and routing AI across your own hardware, approved machines and frontier APIs.
Is Gemini safe for work?
Is Gemini safe for work? A dated, sourced table of what Google does with company data in the Gemini app, Workspace and the Gemini API, and a verdict for each.
What is sovereign AI? The UK position and who offers it
Sovereign AI explained: data, operational and model sovereignty, the UK government's dated funding and compute commitments, and who offers it in the UK.
Self-hosted AI: what it is, what it costs and how to size it
Self-hosted AI explained: laptop, server or private cloud, the software layers, sizing numbers and a worked cost example against API prices, with dated sources.
Claude Code + DeepSeek: setup, limits and where your code goes
Run Claude Code on DeepSeek: the base URL, variables and model names from DeepSeek's docs, what breaks, where prompts are stored and the open-weight path.
Claude Code + OpenRouter: setup, model routing and what happens to your code
Run Claude Code through OpenRouter's Anthropic-compatible endpoint: the three variables, model overrides, how to check the route, and where your prompts go.
Codex + Ollama: run OpenAI Codex CLI on a local model
Run OpenAI Codex CLI on a local model with Ollama: ollama launch codex, the --oss flag, the config.toml provider block, context length, models and limits.
AI tools at work: is each one safe for company data?
AI tools at work, one dated guide per tool: ChatGPT, Claude, Copilot, Gemini, DeepSeek and more. A verdict per plan, plus shadow AI and evaluation guides.
Is ChatGPT safe for work?
Is ChatGPT safe for work? A dated, sourced table of what OpenAI does with company data on each plan, a verdict per plan, and the settings to change.
OpenCode + Ollama: run OpenCode on a local model
Run OpenCode on a local model with Ollama: the one-command launch, automatic discovery in OpenCode 2, the opencode.json provider block, models and limits.
Shadow AI: what it is, how to find it, and an AI acceptable use policy you can copy
Shadow AI is staff using AI tools the organisation has not approved. Dated UK and global figures, how to find it, and a copyable AI acceptable use policy.
Is Claude safe for work?
Is Claude safe for work? A dated, sourced table of how Anthropic handles company data on Free, Pro, Max, Team, Enterprise and the API, and a verdict per plan.
Is Copilot safe for work?
Is Copilot safe for work? A dated, sourced verdict for Microsoft Copilot and GitHub Copilot by plan: training, retention, UK and EU processing, audit, prices.
Is DeepSeek safe for work?
Is DeepSeek safe for work? A dated, sourced verdict for the DeepSeek app, the API and the open weights, with the privacy terms, the bans and the 2025 leak.
Claude Code + Ollama: run Claude Code on your own models
Run Claude Code on a local model with Ollama: one command, the manual environment variables, the models that work, the limits, and the same setup for a whole team.
Local AI for a development team: three setups, sizing and rollout
How to give a whole development team local AI for coding: a model on every laptop, one shared server, or a routed mix. Costs, failure modes and sizing numbers.
Best local LLMs for coding: 10 models by memory tier
The best local LLM for coding at 16, 24, 32, 48 and 64 GB: 10 current models with licence, size, context, tool calling, a dated benchmark and the Ollama tag.
AI coding on your own models: every app and model pairing
Run Claude Code, OpenCode and Codex on models you choose. A table of every pairing guide: the app, the model source, local or cloud, and a link to the setup.
Our security team has blocked public AI tools. What should we evaluate?
Public AI tools blocked by security? Compare enterprise tiers, a private workspace on approved cloud, self-hosted models and a routed mix, with a checklist.
Use cloud AI models with a local model as backup
Use your chosen cloud AI model or endpoint through Aventail™ Link, prepare a local model as backup, and verify where requests run.
What is routing?
AI routing chooses where a model request runs. Learn how Aventail™ connects cloud models, your own computer and another approved device, with optional backups.
Gemini 4 Argon vs Carbon: which should your team use?
Compare Gemini 4 Argon and Carbon for business AI: reasoning, coding, cybersecurity benchmarks, token costs and choosing the right model for each workflow.
Peer-to-peer AI routing: use models on another computer
Use a model on another computer running Aventail Link. Add a same-account, same-network device as a backup and verify where requests run.
Air-gapped AI: what it is, how to deploy it and how to evaluate it
Air-gapped AI runs models and their whole workflow with no link to outside networks. What it means, how to deploy and test it, and a checklist for vendors.
Is Claude Code safe for work?
Is Claude Code safe for work? A dated table of what Anthropic stores by plan, API key, Bedrock and gateway, plus telemetry, permissions and a rollout checklist.
OpenCode vs Claude Code
OpenCode vs Claude Code compared on licence, price, models, local model support, permissions, enterprise controls and telemetry, plus a pilot checklist.
LLM gateway vs inference engine: what does your enterprise need?
Compare LLM gateways, inference engines and fleet management. Choose an architecture for private models, approved cloud APIs and policy-based routing.
Use open models in Claude with Aventail™
Use supported open-weight models in Claude Desktop and Claude Code through Aventail™ Connect, with model requests under your routing rules.
How to use OpenCode with Aventail™ Connect
Connect OpenCode to Aventail™ Link in Connected Apps: no provider config or API key, model chosen in Link, and how to check where requests really ran.
Deploying a private AI workspace across laptops and servers
Deploy a private AI workspace across laptops and servers with Aventail. Plan self-hosted and hybrid setups, staged rollout and governance checks.
How to route between local models and approved cloud APIs
Route AI requests between local open-weight models, peer machines and approved cloud APIs: rules, fallback hops and how to verify where requests ran.
How to reduce AI provider dependency
Keep a tested choice between approved frontier APIs and supported models on your own hardware, without assuming models are interchangeable.
LM Studio vs Aventail™: local AI, devices and routing for teams
Compare LM Studio and Aventail™ for local AI, remote devices and enterprise routing. Assess deployment, data boundaries, cost and beta routing with a practical checklist.
Ollama vs Aventail™: apps, devices and routing
Compare Ollama and Aventail™ for local model serving, API access, app setup and device management. Understand where inference runs and how routing fits.
Keeping sensitive AI requests within approved infrastructure
Decide where each AI request may run, evaluate it on the device, and keep every permitted fallback within the same boundaries.
From GPU capacity to a usable inference service
Give customers a way to deploy models and use your compute, with an inference engine and orchestration layer working together.