Skip to main content
Available now on ModelsLab · Language Model

Sao10K: Llama 3 8B LunarisCreative reasoning meets dialogue

Try Sao10K: Llama 3 8B LunarisAPI Documentation

Merge-optimized generalist. Deploy now.

Strategic Merge

Five models unified

Combines five Llama 3 variants for balanced creativity, reasoning, and general knowledge.

Roleplay Optimized

Dialogue generation excellence

Engineered for creative conversations with minimal cross-context repetition and stylistic diversity.

Efficient Scale

8B parameters, full capability

Lightweight 8-billion parameter model with 8K native context and 128K extended support.

Examples

See what Sao10K: Llama 3 8B Lunaris can create

Copy any prompt below and try it yourself in the playground.

Character Development

“Create a detailed character profile for a seasoned detective in a noir mystery. Include personality traits, background, motivations, and speaking style. Make the character compelling and three-dimensional.”

Dialogue Scene

“Write a tense negotiation dialogue between a diplomat and a rebel leader. Include subtext, pauses, and emotional beats. Make each character's perspective clear and authentic.”

Worldbuilding

“Describe a fictional city's architecture, culture, economy, and governance system. Include specific details about districts, landmarks, and how citizens interact with their environment.”

Story Continuation

“Continue this narrative with vivid sensory details and character development: 'The train pulled into the station at midnight, and everything changed.' Write 500+ words maintaining tone and momentum.”

For Developers

A few lines of code.
Creative dialogue. Three lines.

ModelsLab handles the infrastructure: fast inference, auto-scaling, and a developer-friendly API. No GPU management needed.

  • Serverless: scales to zero, scales to millions
  • Pay per token, no minimums
  • Python and JavaScript SDKs, plus REST API
API Documentation 
import requests


response = requests.post(
    "https://modelslab.com/api/v7/llm/chat/completions",
    json={
  "key": "YOUR_API_KEY",
  "prompt": "",
  "model_id": ""
}
)
print(response.json())

FAQ

Common questions about Sao10K: Llama 3 8B Lunaris

Read the docs

Sao10K Llama 3 8B Lunaris is a strategic merge of five Llama 3-based models designed to balance creativity with improved logical reasoning. It evolves beyond Stheno v3.2 by combining roleplay, storytelling, and general knowledge models into a single 8B parameter generalist.

The model has an 8,192 token native context window with support for up to 128K tokens through extended context, with practical performance around 32K tokens. Max output tokens depend on your API provider configuration.

Pricing is $0.04 per million input tokens and $0.05 per million output tokens. Exact rates may vary by provider, but this model is among the most affordable open-source LLM options available.

It excels at both. The model is optimized for creative roleplay and dialogue generation while maintaining strong general knowledge and reasoning capabilities. It's designed as a versatile generalist that doesn't sacrifice creativity for logic.

Use the Llama 3 Instruct context template with temperature set to 1.4 and min_p at 0.1 for optimal creative output. Adjust temperature lower (0.7-0.9) for more factual or reasoning-focused tasks.

No, Sao10K Llama 3 8B Lunaris is text-only and does not support vision, audio, tool calls, or other modalities. It's a pure language model focused on text generation and dialogue.

Ready to create?

Start generating with Sao10K: Llama 3 8B Lunaris on ModelsLab.

Try Sao10K: Llama 3 8B LunarisAPI Documentation