Builder reference

Model API compatibility at a glance

These rows are intentionally approximate. Always confirm tool-calling, JSON mode, and limits in the provider’s docs for the exact model id you deploy. Use this table to plan migrations and Checks coverage.

ProviderTool callingJSON / structuredContext (approx.)Notes
OpenAInativenativeVaries by model (e.g. 128k–1M class)Responses + Chat Completions; JSON/schema modes differ by endpoint.
AnthropicnativepartialLong context on Sonnet/Opus tiersTool use is first-class; structured output patterns vary by SDK.
Google (Gemini)nativenativeLarge context on 1.5/2 familyVertex vs AI Studio feature parity can differ.
Meta (Llama)variesvariesDepends on host and fine-tuneOpen weights: behavior depends on inference stack and chat template.
MistralnativepartialModel-dependentCheck function-calling docs for your specific model id.
Azure OpenAInativenativeMatches OpenAI model deploymentEnterprise controls, regions, and content filters affect routing.
CoherenativepartialModel-dependentCommand R+ and tool-use APIs differ from embed/classify endpoints.
PerplexityvariesvariesProduct-dependent (API vs consumer)Confirm structured output and tools on the exact API product you use.