# InternLumina-U2

> A Multi-Codebook Diffusion Large Language Model for Omni-Visual Understanding, Image Generation and Editing.

- **Organisation:** internlm
- **Category:** Omni / Any-to-Any
- **Parameters:** 16B
- **Licence:** Apache 2.0 (Commercial use OK)
- **Released:** 2026-09-01
- **Hugging Face:** 0 downloads (30 days), 23 likes

## Estimated memory

| Quant | ≈ GB |
|---|---|
| Q2_K | 7.09 |
| Q3_K_M | 9.59 |
| Q4_K_M | 11.43 |
| Q5_K_M | 13.08 |
| Q6_K | 14.84 |
| Q8_0 | 18.57 |
| F16 | 33.24 |

## Runs with

- [Transformers](https://github.com/huggingface/transformers)

## About

A Multi-Codebook Diffusion Large Language Model for Omni-Visual Understanding, Image Generation and Editing.

📑 Paper · 🌐 Project Page · 💻 Code

Intern Lumina U2 is a unified multimodal model that brings language, image, video and 3D into a single framework, covering text QA, text-to-image generation, image understanding, image editing, video understanding and 3D understanding with one model. It is a 16B-parameter MoE with 1B active parameters (16B-A1B), pairing an efficient sparse backbone with an 8-codebook fully-discrete visual representation built on AToken.

This repo hosts checkpoints of the same architecture trained on different hardware stacks:

• ascend/ — trained on Huawei Ascend NPUs.
• nvidia/ — trained on NVIDIA GPUs.

## Alternatives

- [google/gemma-4-12B-it](https://selfhost.directory/models/google/gemma-4-12b-it) (12B, Apache 2.0, ≈10.59 GB at Q4): Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants.
- [google/gemma-4-E4B-it](https://selfhost.directory/models/google/gemma-4-e4b-it) (8B, Apache 2.0, ≈5.9 GB at Q4): Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants.
- [sensenova/SenseNova-U1.5-8B-MoT](https://selfhost.directory/models/sensenova/sensenova-u1.5-8b-mot) (17.5B, Apache 2.0, ≈12.21 GB at Q4): SenseNova-U1.5-8B-MoT is our latest native unified multimodal checkpoint for more accurate, consistent, reliable, and aesthetically compelling visual creation.
- [google/gemma-4-12B](https://selfhost.directory/models/google/gemma-4-12b) (12B, Apache 2.0, ≈10.59 GB at Q4): Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants.
- [inclusionAI/LLaDA2.0-Uni](https://selfhost.directory/models/inclusionai/llada2.0-uni) (16.3B (1.5B active), Apache 2.0, ≈10.49 GB at Q4): LLaDA2.0-Uni: Unifying Multimodal Understanding and Generation with Diffusion Large Language Model
- [nvidia/NLD-Uni](https://selfhost.directory/models/nvidia/nld-uni) (18.3B, Unknown licence, ≈12.39 GB at Q4): NLD-Uni is a unified multimodal model for image understanding and image generation. It combines a diffusion language model with specialized visual components and a mixture-of-transformers design.

More: https://selfhost.directory/models/internlm/internlumina-u2/alternatives

## Runs well on

- [Tesla T4](https://selfhost.directory/models/internlm/internlumina-u2/on/tesla-t4): Q6_K, est. 13 tok/s
- [Arc A770 16GB](https://selfhost.directory/models/internlm/internlumina-u2/on/arc-a770): Q6_K, est. 23 tok/s
- [Radeon RX 7800 XT](https://selfhost.directory/models/internlm/internlumina-u2/on/rx-7800-xt): Q6_K, est. 26 tok/s
- [Tesla P40](https://selfhost.directory/models/internlm/internlumina-u2/on/tesla-p40): Q8_0, est. 11 tok/s
- [Radeon RX 7900 XTX](https://selfhost.directory/models/internlm/internlumina-u2/on/rx-7900-xtx): Q8_0, est. 31 tok/s
- [Instinct MI50 32GB](https://selfhost.directory/models/internlm/internlumina-u2/on/instinct-mi50): Q8_0, est. 33 tok/s

## Similar size

- [google/gemma-4-12B-it](https://selfhost.directory/models/google/gemma-4-12b-it) (12B)
- [google/gemma-4-12B](https://selfhost.directory/models/google/gemma-4-12b) (12B)
- [ByteDance-Seed/BAGEL-7B-MoT](https://selfhost.directory/models/bytedance-seed/bagel-7b-mot) (14.7B)
- [sensenova/SenseNova-U1.5-8B-MoT](https://selfhost.directory/models/sensenova/sensenova-u1.5-8b-mot) (17.5B)
- [HiDream-ai/HiDream-E1-Full](https://selfhost.directory/models/hidream-ai/hidream-e1-full) (17.1B)
- [inclusionAI/LLaDA2.0-Uni](https://selfhost.directory/models/inclusionai/llada2.0-uni) (16.3B)

## FAQ

**How much VRAM does InternLumina-U2 need?**
About 11 GB at Q4_K_M with an 8K context. Q8_0 needs about 19 GB and full precision about 33 GB. Longer contexts need more memory for the KV cache.

**Can InternLumina-U2 run on the Tesla T4?**
Yes. It fits at Q6_K (about 15 GB), at an estimated 13 tokens/s. It fits comfortably on the Arc A770 16GB, Radeon RX 7800 XT, Tesla P40.

**Can I use InternLumina-U2 commercially?**
Yes. It is released under Apache 2.0. Permissive: use, modify and sell freely; keep the notice and licence text.

## Links

- huggingface: https://huggingface.co/internlm/InternLumina-U2
- homepage: https://internlm.github.io/InternLumina-U2/
- github: https://github.com/InternLM/InternLumina-U2
- paper: https://arxiv.org/abs/2610.20818

---
Source: https://selfhost.directory/models/internlm/internlumina-u2