Managed Inference and Agents API with Kimi K2.5

Table of Contents [expand]

Last updated February 18, 2026

Kimi K2.5 is a large language model (LLM) from Moonshot AI that supports conversational chat, tool-calling, massive context processing, and parallel agent swarms for complex tasks with dual-mode reasoning. It offers a text-to-text solution provided via Amazon Bedrock that runs on AWS compute in the US region.

  • Model ID: kimi-k2-5
  • Region: us

When to Use This Model

Kimi K2.5 supports various common use cases, including long-document analysis, complex research synthesis, and office automation. It’s optimized for massive 256K context processing, advanced mathematical reasoning, and autonomous “Agent Swarm” workflows that decompose and execute tasks in parallel.

Usage

Kimi K2.5 follows our /v1/chat/completions API schema.

To provision access to the model, attach a Managed Inference and Agents add-on to your app $APP_NAME:

heroku addons:create heroku-inference:standard -a $APP_NAME

Using config variables, you can invoke the model in various ways:

Rate Limits

  • Maximum requests per minute: 150
  • Maximum tokens per minute: 800,000

Prompt Caching

Prompt caching isn’t supported for Kimi K2.5.

Example curl Request

To retrieve and export your API credentials:

export INFERENCE_KEY=$(heroku config:get -a $APP_NAME INFERENCE_KEY)
export INFERENCE_URL=$(heroku config:get -a $APP_NAME INFERENCE_URL)

Text to Text

curl $INFERENCE_URL/v1/chat/completions \
-H "Authorization: Bearer $INFERENCE_KEY" \
-d @- <<EOF
{
  "model": "kimi-k2-5",
  "messages": [
    { "role": "user", "content": "Hello!" },
    { "role": "assistant", "content": "Hi there! How can I assist you today?" },
    { "role": "user", "content": "What's the weather like in Portland, Oregon right now?" }
  ],
  "temperature": 0.5,
  "max_tokens": 100,
  "stream": false,
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "get_weather",
        "description": "Fetches the current weather for a given city.",
        "parameters": {
          "type": "object",
          "properties": {
            "city": {
              "type": "string",
              "description": "The name of the city to get weather for."
            }
          },
          "required": ["city"]
        }
      }
    }
  ],
  "tool_choice": "auto",
  "top_p": 0.9
}
EOF