Chat Models Comparison

This table compares various Chat Models supported by Spring AI, detailing their capabilities:

  • Multimodality: The types of input the model can process (e.g., text, image, audio, video).

  • Tools/Function Calling: Whether the model supports function calling or tool use.

  • Streaming: If the model offers streaming responses.

  • Retry: Support for retry mechanisms.

  • Observability: Features for monitoring and debugging.

  • Built-in JSON: Native support for JSON output.

  • Local deployment: Whether the model can be run locally.

  • OpenAI API Compatibility: If the model is compatible with OpenAI’s API.

Provider Multimodality Tools/Functions Streaming Retry Observability Built-in JSON Local OpenAI API Compatible

Anthropic Claude

text, pdf, image

yes
yes
yes
yes
no
no
no

DeepSeek (OpenAI-proxy)

text

yes
yes
yes
yes
yes
yes
yes

Google GenAI

text, pdf, image, audio, video

yes
yes
yes
yes
yes
no
no

Groq (OpenAI-proxy)

text, image

yes
yes
yes
yes
no
no
yes

Mistral AI

text, image, audio

yes
yes
yes
yes
yes
no
yes

MiniMax (Anthropic-proxy)

text, image

yes
yes
yes
yes
no
no
yes

NVIDIA (OpenAI-proxy)

text, image

yes
yes
yes
yes
no
no
yes

Ollama

text, image

yes
yes
yes
yes
yes
yes
yes

OpenAI

In: text, image, audio Out: text, audio

yes
yes
yes
yes
yes
no
yes

Perplexity (OpenAI-proxy)

text

no
yes
yes
yes
no
no
yes

Amazon Bedrock Converse

text, image, video, docs (pdf, html, md, docx …​)

yes
yes
yes
yes
no
no
no