How to Use Amazon Nova Micro on Composite
At a glance
| Model ID | amazon/nova-micro-v1 |
|---|---|
| Free ID | amazon/nova-micro-v1:free |
| Made by | Amazon |
| Context window | 128,000 tokens |
| Longest reply | 5,120 tokens |
| Input | Text |
| Reasoning | No |
| Tool calling | Yes |
Amazon Nova Micro 1.0 is a text-only model that Amazon describes as having the lowest latency in the Nova family. It has a 128,000-token window and writes replies of up to 5,120 tokens.
It is moderated
Nova Micro is the one model on the free list whose provider moderates requests. Prompts or replies that the provider's filter rejects come back as errors instead of text. Mature roleplay and other content the filter blocks will not run on it, whatever preset you use. Pick another model for those chats.
Set a reply limit it accepts
Many chat presets ask for a large maximum reply length. Nova Micro stops at 5,120 tokens, so set your app's response length to that or less. A typical chat reply is a few hundred tokens, well inside the cap.
A short list of settings
It accepts temperature, top P, top K, max tokens, stop sequences and tools. Repetition, frequency and presence penalties and min P are not supported, so presets that rely on them will not have the effect they have on other models.
What to use it for
Short, fast, safe-for-work tasks: classifying a message, extracting fields from text, answering a quick question, or acting as a small tool-calling step in a script. It supports tool calling, so it can sit inside an agent loop that needs a quick decision rather than long writing.
Connect it to your chat app
- Create an API key on the API Keys page.
- In SillyTavern, ChubAI, Agnai or RisuAI, choose an OpenAI-compatible provider and set the base URL to
https://composite.lucidity.sh/v1. JanitorAI wants the full URL:https://composite.lucidity.sh/v1/chat/completions. - Paste the key, then type
amazon/nova-micro-v1:freeas the model name. - Send a short message to check the connection before you load a long chat.
Step-by-step setup for each app is in the JanitorAI and SillyTavern guides, and the docs hub covers the others.
From code, it is a normal OpenAI-compatible request:
curl https://composite.lucidity.sh/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "amazon/nova-micro-v1:free", "messages": [{"role": "user", "content": "Hello"}]}'Free access (as of October 21, 2026)
On October 21, 2026, this model is on Composite's free list as amazon/nova-micro-v1:free. The plain amazon/nova-micro-v1 stays paid. Every account shares 80 requests a day across all of the free models, separate from Synth's count, and the count resets at midnight UTC. Credits and plans do not raise it.
The free list changes. Check the free models on the Models page for the current set, and the features guide for the rules. A :free id that is no longer on the list returns a model-not-found error and charges nothing. When the 80 requests are used up, requests stop with an error until the reset; Composite never charges your credits to continue. You can switch to amazon/nova-micro-v1 on credits or a plan, or to Synth.