MOSS-VL-Realtime-GGUF by Skycontact: Open-Weight Model

Model Card

By Skycontact, published under apache-2.0, revision dca2156a7a19.

GGUF quantization of OpenMOSS-Team/MOSS-VL-Realtime-NF4 (11B cross-attention VLM: Qwen3 text trunk with 12 pure cross-attn layers + SigLIP tower with deepstack), repackaged for the SkyVision.cpp inference engine. - NF4 weights are dequantized bit-exactly against bitsandbytes CUDA first (checkmossvlnf4.py); - the fused cross-KV projection weight is built at load time (engine-side); 12 GB GPU deployments: --n-ctx 4096 (and MOSSXCTX=6144 for long videos). Full chat, multi-turn sessions, realtime streaming (proactive narration with ) and a web UI are available via --serve. Follows the base model's license (see the OpenMOSS-Team/MOSS-VL-Realtime-NF4 page).

Read Skycontact's full model card

Moss-VL-Realtime GGUF (Q4_K)

GGUF quantization of OpenMOSS-Team/MOSS-VL-Realtime-NF4 (11B cross-attention VLM: Qwen3 text trunk with 12 pure cross-attn layers + SigLIP tower with deepstack), repackaged for the SkyVision.cpp inference engine.

Built for SkyVision.cpp only. The GGUF declares general.architecture = moss_vl; llama.cpp / ollama / LM Studio and other generic GGUF runtimes cannot load it. This repository is not part of the generic GGUF ecosystem - do not use it as a signal of whether the model "works".

Files

File	Size	Notes
`Moss-VL-Realtime-q4k.gguf`	8.4 GB	NF4 → Q4_K (linears) + Q6_K (norm-adjacent) + F16 (vision), fits a 12 GB GPU

Quantization scheme

NF4 weights are dequantized bit-exactly against bitsandbytes CUDA first (check_moss_vl_nf4.py);
then requantized: cross-attn / norm-adjacent linears → Q6_K, the rest → Q4_K, vision tower + 1D tensors kept F16/F32;
the fused cross-KV projection weight is built at load time (engine-side);
verified against the official stack: image prefill token stream 24/24 and video 78/78 token-exact, layer-2 cross-attention internals corr ≥ 0.999, KV-cache readback vs official roped K corr 0.9997.

Usage

vqa-cli --model Moss-VL-Realtime-q4k.gguf --image photo.jpg --ask "Describe this image."
vqa-cli --model Moss-VL-Realtime-q4k.gguf --video clip.mp4 --max-frames 8 --ask "What happens?"

12 GB GPU deployments: --n-ctx 4096 (and MOSS_XCTX=6144 for long videos). Full chat, multi-turn sessions, realtime streaming (proactive narration with <|silence|>) and a web UI are available via --serve.

License

Follows the base model's license (see the OpenMOSS-Team/MOSS-VL-Realtime-NF4 page).

Identity and Version

Repository: SkyContact/MOSS-VL-Realtime-GGUF
Publisher: Skycontact
Task: Video text to text
Modality: Other
Library: SkyVision.cpp
Parameters: Not stated by the source
Languages: vlm
Revision: dca2156a7a19c3daa4841c368bc8944e60d20b4f
First published: 2026-10-17
Last updated: 2026-10-17

Files and Weights

3 files, 8.4 GB in total. The weights are 1 file totalling 8.4 GB in gguf.

Weights1 file · 8.4 GB

Documentation1 file · 2.2 KB

Repository1 file · 1.6 KB

Every file

File	Type	Size	SHA-256
Moss-VL-Realtime-q4k.gguf	Weights	8.4 GB	605470f81a2e
README.md	Documentation	2.2 KB	—
.gitattributes	Repository	1.6 KB	—

License and Download

License: apache-2.0
Access: Open weights, no gate
Download size: 8.4 GB

Download from Skycontact

Released by Skycontact through its official repository on Hugging Face. Read the license.

Built From

Derived from OpenMOSS-Team/MOSS-VL-Realtime-NF4
Quantized from OpenMOSS-Team/MOSS-VL-Realtime-NF4

Memory Requirements

Precision	Weights in memory
As published	8.4 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About MOSS-VL-Realtime-GGUF

Can I use MOSS-VL-Realtime-GGUF commercially?

Yes. MOSS-VL-Realtime-GGUF is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

Similar Models

Model · Video text to text

VideoGuard-Qwen3.5-4B-Safety-RL-Uncensored-GGUF

Prithiv Sakthi

C/C++ — https://github.com/ggml-org/llama.cpp

Open weights apache-2.0 transformers

View model

Model · Video text to text

VideoGuard-Qwen3.5-9B-Safety-RL-Uncensored-GGUF

Prithiv Sakthi

C/C++ — https://github.com/ggml-org/llama.cpp

Open weights apache-2.0 transformers

View model

Model · Video text to text

EventLens-4B

Guowen Zhang

EventLens-4B is a video-language model developed through event-structure reinforcement learning and multi-teacher on-policy distillation (MT-OPD). Built on Qwen3-VL-4B-Instruct, it learns to recognize event boundaries, progression, and relations, alongside temporal grounding and video question answering. The model is part of EventLens: Event-Structure Reinforcement Learning for Video Understanding, developed by AIM3 Lab, Renmin University of China, and Kling AI. Selected results reported in the EventLens paper, compared with the corresponding pretrained backbone under the shared 256-frame protocol. QVHighlights, ActivityNet, and Charades-STA use mIoU. E.T. Bench and VinoGround use their…

Open weights cc-by-nc-sa-4.0 4.8B parameters 262,144 tokens transformers

View model

Model · Video text to text

EventLens-8B

Guowen Zhang

EventLens-8B is a video-language model developed through event-structure reinforcement learning and multi-teacher on-policy distillation (MT-OPD). Built on Qwen3-VL-8B-Instruct, it learns to recognize event boundaries, progression, and relations, alongside temporal grounding and video question answering. The model is part of EventLens: Event-Structure Reinforcement Learning for Video Understanding, developed by AIM3 Lab, Renmin University of China, and Kling AI. Selected results reported in the EventLens paper, compared with the corresponding pretrained backbone under the shared 256-frame protocol. QVHighlights, ActivityNet, and Charades-STA use mIoU. E.T. Bench and VinoGround use their…

Open weights cc-by-nc-sa-4.0 8.8B parameters 262,144 tokens transformers

View model