Compatibility

Compatibility facts for API, SDK, MCP, Agent, RAG, and framework surfaces.

SourceLayerTargetStatusEvidenceUpdated
claude-opus-4-1

Claude Opus 4.1 is modeled around Anthropic Messages API compatibility.

api anthropic-messages-api supported Anthropic docs and TypeScript SDK links are attached to Claude model entries. 2026-05-18
claude-haiku-3-5

Claude Haiku 3.5 is used through Anthropic's Messages API and official SDKs.

api Anthropic Messages API supported Anthropic's model overview lists Claude Haiku 3.5 model identifiers, and the Anthropic TypeScript SDK documents server-side Claude API usage. 2026-05-19
claude-sonnet

Claude Sonnet is modeled around Anthropic Messages API compatibility for agent planning and code review workflows.

api anthropic-messages-api supported Claude Sonnet entries include Anthropic model documentation and Anthropic TypeScript SDK GitHub sources. 2026-05-18
deepseek-v4

DeepSeek V4 is modeled as usable through OpenAI-compatible SDK patterns when current DeepSeek API behavior is verified.

sdk openai-compatible-sdk verify_required DeepSeek V4 uses an OpenAI-compatible API style in the model entry and links to DeepSeek official API documentation. 2026-05-18
devstral-2

Devstral 2 should be integrated through Mistral API and official SDK paths after verifying the current model identifier.

sdk Mistral SDK supported Mistral model docs list Devstral 2 for code-agent workflows, and the Mistral TypeScript client documents official API integration. 2026-05-19
gemini-2-5-flash

Gemini 2.5 Flash is accessible through the Google Gen AI SDK for current Gemini API and Vertex AI integration paths.

sdk Google Gen AI SDK supported Google's js-genai repository documents @google/genai as the current TypeScript and JavaScript SDK for Gemini and Vertex AI. 2026-05-19
gemini-3-5-flash

Gemini 3.5 Flash is accessible through the Google Gen AI SDK for Gemini API and Vertex AI integration.

sdk Google Gen AI SDK supported Google's js-genai repository documents @google/genai as the current TypeScript and JavaScript SDK for Gemini and Vertex AI model access. 2026-05-21
gemini-3-pro

Gemini model entries are modeled as compatible with Google Gen AI SDK workflows.

sdk google-gen-ai-sdk supported Gemini model entries include Google Gemini API docs and Google Gen AI SDK GitHub links. 2026-05-18
gemini-pro

Gemini Pro is modeled as a legacy Gemini entry that should be verified against current Google Gen AI SDK support before production use.

sdk google-gen-ai-sdk verify_required Gemini Pro includes Google Gemini API documentation and Google Gen AI SDK GitHub sources. 2026-05-18
gpt-5-2-codex

GPT-5.2 Codex is modeled as a coding-optimized OpenAI entry that should be verified against current Responses API availability and limits.

api openai-responses-api verify_required GPT-5.2 Codex includes OpenAI model documentation and OpenAI SDK GitHub sources. 2026-05-18
gpt-oss-120b

gpt-oss-120b can be evaluated through OpenAI's gpt-oss reference implementations and Responses-compatible examples.

framework gpt-oss reference stack supported The OpenAI gpt-oss repository documents reference implementations, harmony format guidance, and Responses-compatible examples for the gpt-oss family. 2026-05-19
gpt-oss-20b

gpt-oss-20b can be evaluated through local runtimes such as Ollama, but production compatibility depends on serving configuration and prompt format support.

framework Ollama local runtime adapter_required The OpenAI gpt-oss repository includes local usage paths for gpt-oss through Ollama and LM Studio while emphasizing format requirements. 2026-05-19
grok-4-3

Grok 4.3 is modeled as compatible with xAI Responses API workflows and OpenAI-compatible client usage through the xAI base URL.

api xAI Responses API supported xAI documentation includes Responses API examples for Grok 4.3 and the xAI SDK repository documents client integration. 2026-05-18
llama-4-maverick

Llama 4 Maverick is modeled as compatible with Transformers-based inference when license approval, hardware, and package versions are satisfied.

framework Transformers supported Meta's model card includes Transformers usage guidance and the meta-llama GitHub repository provides Llama 4 tooling notes. 2026-05-18
mistral-large-3

Mistral Large 3 is accessible through the official Mistral SDK for TypeScript, Python, and other languages.

sdk Mistral SDK supported Mistral AI documentation and SDK repositories confirm SDK support for Mistral Large 3 hosted API and self-hosted deployments. 2026-05-21
mistral-medium-3-5

Mistral Medium 3.5 is modeled as compatible with Mistral SDK workflows once current model availability is verified.

sdk mistral-sdk verify_required Mistral Medium 3.5 includes Mistral official documentation and Mistral SDK GitHub sources. 2026-05-18
o3

OpenAI o3 and o4-mini use the Responses API for reasoning, chain-of-thought, and structured output tasks.

api OpenAI Responses API supported OpenAI API documentation lists o-series models under the Responses API with reasoning effort parameters and vision support. 2026-05-21
deepseek-v3-1

DeepSeek API entries are modeled as compatible with OpenAI-style SDK integration patterns when official API behavior is verified.

sdk openai-compatible-sdk supported DeepSeek model entries include OpenAI-compatible API style and GitHub model-family references. 2026-05-18
gpt-5-2

GPT-5.2 is modeled as using the OpenAI Responses API for tool-using and agentic workflows.

api openai-responses-api supported OpenAI API reference describes the Responses endpoint as the interface for model responses with text, image, tool, and stateful interaction support. 2026-05-18
qwen3

Qwen3 can fit OpenAI-compatible integration patterns when served through frameworks such as SGLang or vLLM.

framework OpenAI-compatible serving adapter_required Qwen3 GitHub documentation describes deployment through SGLang and OpenAI-compatible API service patterns. 2026-05-18