Last updated January 15, 2026
The Heroku Managed Inference and Agents API offers an easy way to access various large foundational AI models. These include supported language (chat), embedding, and diffusion (image) models.
To learn more about the API endpoints, including parameters, usage examples, and responses, view the documentation linked below.
| Endpoint Documentation |
Supported Models |
Example Code |
| Chat Completions |
Claude Opus 4.5, Claude 4.5 Sonnet, Claude 4.5 Haiku, Nova 2 Lite, Kimi K2 Thinking, MiniMax M2, Qwen3-Coder-480B, Qwen3-235B, Claude 4 Sonnet, Claude 3.7 Sonnet, Claude 3.5 Sonnet Latest, Claude 3.5 Haiku, Claude 3 Haiku, Amazon Nova Lite, Amazon Nova Pro, OpenAI gpt-oss-120b |
Python, JavaScript, Ruby |
| MCP Servers |
Claude Opus 4.5, Claude 4.5 Sonnet, Claude 4.5 Haiku, Nova 2 Lite, Kimi K2 Thinking, MiniMax M2, Qwen3-Coder-480B, Qwen3-235B, Claude 4 Sonnet, Claude 3.7 Sonnet, Claude 3.5 Sonnet Latest, Claude 3.5 Haiku, Claude 3 Haiku, Amazon Nova Lite, Amazon Nova Pro, OpenAI gpt-oss-120b |
|
| Agents (Heroku) |
Claude Opus 4.5, Claude 4.5 Sonnet, Claude 4.5 Haiku, Nova 2 Lite, Kimi K2 Thinking, MiniMax M2, Qwen3-Coder-480B, Qwen3-235B, Claude 4 Sonnet, Claude 3.7 Sonnet, Claude 3.5 Sonnet Latest, Claude 3.5 Haiku, Claude 3 Haiku, Amazon Nova Lite, Amazon Nova Pro, OpenAI gpt-oss-120b |
|
| Embeddings |
Cohere Embed Multilingual |
Python, JavaScript, Ruby |
| Image Generations |
Stable Image Ultra |
Python, JavaScript, Ruby |
| Reranking |
Cohere Rerank 3.5, Amazon Rerank 1.0 |
|