Live tool

LLM Model Selection Wizard

Answer a few questions about latency, context window, modality, task complexity, cost ceiling, and data residency, and get a recommended LLM across Azure AI Foundry, Vertex (Gemini), and Bedrock — with the reasoning behind it and ranked alternatives. Runs entirely in your browser.

Live Cloud Wizard LLM Azure OpenAI Gemini Bedrock Model Selection
Open tool
Catalog note: Model specs, output pricing, and EU availability are illustrative, last verified 1 Jun 2026. Cross-cloud LLM details change fast — verify against each provider before committing. The wizard ranks on requirement fit and cost, not on quality benchmarks.

Your requirements

The recommendation updates live as you change these.

Recommendation

Top fit plus alternatives, with the reasoning behind each.

How the wizard decides

Hard filters first, then preferences.

Hard filters (disqualify) Modality, context window, cloud, EU residency, and any cost ceiling. A model that fails one is removed entirely — no amount of other strengths brings it back.
Soft preferences (rank) Latency priority and task complexity drive most of the ranking; output cost is a bounded secondary signal, so a cheap model never wins on price alone for a complex task.
Ties Equal-fit models break ties by lower output cost, then by a stable order — so the same inputs always give the same answer.

What this does not cover

Check these before you commit.

1. Quality benchmarks Capability tiers are coarse. Run your own eval on your actual task — a "balanced" model can beat a "frontier" one on a narrow domain.
2. Rate limits & quotas Provisioned throughput, regional quotas, and per-model limits can rule out a pick the scoring would otherwise favour.
3. Exact pricing & regions Output cost here is representative. Confirm input/cached pricing and the specific region you need.
4. Fine-tuning & features Fine-tuning availability, structured outputs, and tool-calling support vary by model and aren't part of the ranking.