hostinglinks.
PRACTICAL TECH, EXPLAINED
HostingLinks / Explore

MODELS THE PROVIDER RUNS FOR YOU

Claude, GPT and Gemini. Kept separate from local AI.

Hosted models do not have downloadable weights you can size with a VRAM calculator. Compare their role, access and provider instead. This catalogue is reviewed regularly because hosted lineups change quickly.

No local VRAM requiredThe provider runs the model infrastructure.
Model names can change quicklyEach entry carries a review date and official source.
Local is a separate questionOpen-weight families live in the local model catalogue.
Anthropic

Claude current family

Frontier / specialist

Claude Opus 5.5

Complex work requiring careful judgement and demanding agentic tasks.

Access: Claude and Anthropic API

Reviewed 14 October 2026

Official source →
Current general model

Claude Sonnet 5.5

Everyday knowledge work, coding, documents, slides and spreadsheets.

Access: Claude and Anthropic API

Reviewed 14 October 2026

Official source →
Fast / efficient

Claude Haiku 5.5

High-volume, cost-sensitive and speed-sensitive work.

Access: Claude and Anthropic API

Reviewed 14 October 2026

Official source →
OpenAI

GPT / ChatGPT current family

Frontier / specialist

GPT-6 Astra

Frontier computer use, software engineering, science and difficult professional work.

Access: ChatGPT/API availability depends on plan and rollout

Reviewed 14 October 2026

Official source →
Current general model

GPT-6.1 Sol

High-capability coding, computer use and professional work with lower cost than Astra.

Access: OpenAI API and supported OpenAI products

Reviewed 14 October 2026

Official source →
Fast / efficient

GPT-6 Luna

Faster everyday GPT-6-family work and structured decision workloads.

Access: ChatGPT/API availability depends on product

Reviewed 14 October 2026

Official source →
Google

Gemini current family

Frontier / specialist

Gemini 4 Argon

Deep multi-step work across coding, finance, legal, cybersecurity and science.

Access: Gemini app for eligible plans and Gemini API

Reviewed 14 October 2026

Official source →
Fast / efficient

Gemini 3.8 Flash

Current fast Gemini API model for high-throughput general work.

Access: Gemini API and Google AI Studio

Reviewed 14 October 2026

Official source →
Current general model

Gemini 3.6 Flash

General fast hosted model and replacement for the deprecated Gemini 3.5 Flash endpoint.

Access: Gemini API and Google AI Studio

Reviewed 14 October 2026

Official source →