How to Use Nex-N2.5-Mini on Composite

At a glance

Nex-N2.5 is Nex AGI's agentic model family, built, in Nex AGI's words, to turn goals into working, verified outcomes. Its stated core strength is agentic coding within a visual feedback loop, such as exploring a codebase and implementing changes across several files. Mini is the smaller model in the family. It reads text and images and has a 262,144-token window.

Check your settings first

This model accepts fewer settings than most of the free list. On Composite it takes temperature, top P, top K, max tokens, reasoning and structured output. It does not list stop sequences, min P, or repetition, frequency or presence penalties. A roleplay preset built around those will still send them, but they will not shape the output, so judge the model on temperature and top P alone.

Tool calling is not in the model's supported settings on this route. If you plan to use it in a coding agent that relies on OpenAI-style tool calls, test one tool call before you start a long task.

Reasoning efforts: medium or high

Reasoning is optional, and the efforts are medium and high. There is no low setting, so nex-agi/nex-n2.5-mini:free[low] will not work. Use [medium] for everyday prompts and [high] for planning or debugging, or leave the suffix off to use the model's default.

When to choose it

Choose Nex-N2.5-Mini when the job is building or fixing something and you can give it a screenshot or image of the result to check against. For long story chats, models that accept a full sampler set, such as Mistral Nemo or DeepSeek V4 Flash 0731, give a preset more to work with.

Connect it to your chat app

1. Create an API key on the API Keys page.
2. 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.
3. Paste the key, then type nex-agi/nex-n2.5-mini:free as the model name.
4. 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": "nex-agi/nex-n2.5-mini: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 nex-agi/nex-n2.5-mini:free. The plain nex-agi/nex-n2.5-mini 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 nex-agi/nex-n2.5-mini on credits or a plan, or to Synth.

More

• Model page with live pricing and rankings
• Model modifiers, including reasoning effort suffixes
• Best LLMs for Roleplay in 2026
• All model guides