# inclusionAI Releases Ling-3.0-tiny, an MIT-Licensed MoE

> The new hybrid mixture-of-experts model targets efficient text generation under a permissive license.

Published by The Open Weights on Aug 10, 2026. Canonical: https://theopenweights.com/news/ling-3-0-tiny-gl7r

## Key facts

- Company: inclusionAI
- Model: Ling-3.0-tiny
- Version: 3.0-tiny
- Category: Text / LLM
- Modalities: Text / LLM
- License: MIT (Open weights, commercial use allowed)
- Context window: 131K tokens
- Architecture: BailingMoeV3ForCausalLM
- Precision: BF16
- Weights size: 15.8 GB
- Significance: notable
- Published: 2026-08-10
- Last verified: 2026-08-24
- Hugging Face: https://huggingface.co/inclusionAI/Ling-3.0-tiny
- Canonical URL: https://theopenweights.com/news/ling-3-0-tiny-gl7r

inclusionAI has published Ling-3.0-tiny, a compact language model built on a hybrid mixture-of-experts (MoE) architecture. The model is available now on [Hugging Face](https://huggingface.co/inclusionAI/Ling-3.0-tiny) and is distributed under the MIT license, one of the most permissive terms in open-source software.

The "tiny" label positions this release as the entry point in the Ling 3.0 family, aimed at developers who want a lighter footprint without abandoning the efficiency gains that MoE designs promise. Mixture-of-experts models activate only a subset of their parameters for any given token, which can deliver strong quality per unit of compute compared with dense models of similar size.

## Why it matters

Permissive licensing remains a decisive factor for teams choosing which open models to build on. MIT terms allow commercial use, modification, and redistribution with minimal restrictions, lowering the barrier for startups and researchers who want to fine-tune or embed the model in their own products.

- Hybrid MoE architecture for efficient inference
- MIT license, suitable for commercial deployment
- Text generation as the primary modality

The release record does not specify parameter counts, context length, or benchmark results, so builders will want to consult the [model card](https://huggingface.co/inclusionAI/Ling-3.0-tiny) directly before committing to production use. As the first public entry in the Ling 3.0 line, it also signals where inclusionAI intends to take the family next.

## Get the model

- [Hugging Face](https://huggingface.co/inclusionAI/Ling-3.0-tiny)

## Sources

- [inclusionAI/Ling-3.0-tiny](https://huggingface.co/inclusionAI/Ling-3.0-tiny) — Hugging Face, Aug 10, 2026

---
Source: The Open Weights (https://theopenweights.com/). Aggregated and written by Claude, curated by humans. Cite as: "inclusionAI Releases Ling-3.0-tiny, an MIT-Licensed MoE", The Open Weights, Aug 10, 2026, https://theopenweights.com/news/ling-3-0-tiny-gl7r