How to Use Ling 3.0 Flash on Composite

At a glance

Ling-3.0-flash is a 124B-parameter mixture-of-experts model from inclusionAI, with about 5.1B parameters active for each token. inclusionAI designed it with token efficiency and production-scale agentic inference as the main priorities. It reads text only, has a 262,144-token window, and writes replies of up to 32,768 tokens.

What the active parameter count tells you

A mixture-of-experts model only runs part of its weights for each token. Ling runs about 5.1B of its 124B, a smaller share than DeepSeek V4 Flash 0731's 13B active out of 284B. The total size sets how much the model can know; the active size is closer to how much computing each token gets. It is one reason to test Ling on your own prompts rather than assume it matches a dense model of either size.

Settings it accepts

Ling takes the full set of chat samplers: temperature, top P, top K, min P, repetition penalty, and frequency and presence penalties, plus stop sequences and logit bias. A SillyTavern or JanitorAI preset will apply as written. It also supports tool calling and a JSON response format, which fits the agent use inclusionAI describes.

The model accepts a reasoning setting, but no effort levels are listed for it. Leave the bracket suffixes off the model ID and control reasoning in the request body or your app's settings.

Reply length

The 32,768-token reply cap is far above what a chat message needs, but it matters for long single outputs such as a full chapter or a large generated file. Ask for the work in parts when a single reply would pass it.

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 inclusionai/ling-3.0-flash: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": "inclusionai/ling-3.0-flash: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 inclusionai/ling-3.0-flash:free. The plain inclusionai/ling-3.0-flash 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 inclusionai/ling-3.0-flash 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