Compatibility Overview
How Agno adapters, provider endpoints, and model IDs determine feature compatibility.
Model compatibility depends on the Agno adapter, the provider endpoint, and the selected model ID. Capabilities still vary across models behind the shared interface.
Shared Model Interface
Every concrete Model adapter implements synchronous, asynchronous, streaming, and non-streaming invocation methods. Request features require support from additional layers:
| Feature | What to verify |
|---|---|
| Streaming | The selected model and endpoint stream responses. |
| Tool calling | The selected model accepts tool schemas and returns tool calls. |
| Structured output | The adapter and model support the selected native schema, JSON mode, or structured-output tool path. |
| Multimodal input and output | The adapter handles the media type and the selected model accepts or returns that modality. |
OpenAI-compatible adapters reuse an API format. Feature parity with OpenAI models varies. Gateways can route one adapter to models with different capabilities.
Multimodal Support
Each checkmark records a media path implemented by the adapter and corroborated by provider documentation, an Agno provider guide, a matching cookbook, or an automated test. A checkmark is an adapter-level signal. Select a model ID that supports the same modality. A blank cell means this page makes no compatibility claim.
| Agno Adapter | Image Input | Audio Input | Audio Responses | Video Input | File Input |
|---|---|---|---|---|---|
AIMLAPI | ✅ | ||||
Claude (Anthropic) | ✅ | ✅ | |||
AwsBedrock | ✅ | ✅ | ✅ | ||
Claude (AWS Bedrock) | ✅ | ||||
AzureAIFoundry | ✅ | ||||
AzureFoundryClaude | ✅ | ✅ | |||
AzureOpenAI | ✅ | ||||
Cerebras | |||||
CerebrasOpenAI | |||||
Cloudflare | |||||
Cohere | ✅ | ||||
CometAPI | ✅ | ||||
DashScope | ✅ | ||||
DeepInfra | |||||
DeepSeek | |||||
Fireworks | |||||
Gemini | ✅ | ✅ | ✅ | ✅ | ✅ |
GeminiInteractions | ✅ | ✅ | ✅ | ✅ | ✅ |
Groq | ✅ | ||||
HuggingFace | |||||
WatsonX | ✅ | ||||
Inception | |||||
InternLM | |||||
LangDB | |||||
LiteLLM | ✅ | ✅ | ✅ | ||
LiteLLMOpenAI | ✅ | ✅ | |||
LlamaCpp | |||||
Llmman | |||||
LMStudio | ✅ | ||||
Llama | ✅ | ||||
LlamaOpenAI (currently blocked by a formatting error) | ✅ | ||||
MiniMax | |||||
MistralChat | ✅ | ||||
MoonShot | |||||
N1N | |||||
Nebius | |||||
Neosantara | |||||
Nexus | |||||
Nvidia | |||||
Ollama | ✅ | ||||
OllamaResponses | |||||
OpenAIChat | ✅ | ✅ | ✅ | ✅ | |
OpenAIResponses | ✅ | ✅ | |||
OpenResponses | |||||
OpenRouter | |||||
OpenRouterResponses | |||||
Perplexity | |||||
Portkey | |||||
RampRouter | |||||
Requesty | |||||
Sambanova | |||||
Siliconflow | |||||
Synthorai | |||||
TokenLab | |||||
Together | ✅ | ||||
TrustedRouter | |||||
TuningEngines | |||||
V0 | ✅ | ||||
VLLM | |||||
Claude (Vertex AI) | ✅ | ✅ | |||
xAI | ✅ | ||||
xAIResponses | |||||
MiMo |