The Open Weights
LatestModelsLeaderboardsCompanies
Subscribe
The Open Weights

The daily record of open-source AI. New model releases, leaderboards, and what's coming next — written for people who ship.

Refreshed every 12 hours

Discover

  • Latest releases
  • New today
  • Trending models

Browse

  • All models
  • Companies
  • Categories
  • Leaderboards

About

  • About
  • Editorial policy
  • Hardware estimates
  • RSS feed
  • llms.txt
  • Newsletter

© 2026 The Open Weights. An independent publication.

PrivacyTermsSMSAggregated by Claude · curated by humans.
LatestKORMo VLWeb-11B
KORMo VLVision-Language

KORMo-Web-11B targets Korean and English GUI agents

An 11-billion-parameter vision-language model built to read screens and drive web and desktop interfaces across two languages.

Nov 14, 2026
UpdateOther
KORMo-Web-11B

A new open vision-language model called KORMo-Web-11B has arrived on Hugging Face, aimed squarely at the growing field of GUI and web automation. At 11 billion parameters, the model is designed to interpret on-screen content and act as an agent that can navigate web pages and desktop interfaces in both Korean and English, according to its model page.

The release lands in a category that has quickly become one of the most competitive corners of multimodal AI: so-called "computer use" models that can look at a screenshot, locate interface elements, and decide where to click or type. Most high-profile entries in this space have been English-first, so a model that treats Korean as a primary language is a notable addition for developers building agents for Korean-language sites and applications.

Why it matters

GUI-grounded agents depend on tight coupling between visual perception and action planning, and language coverage directly shapes how useful they are in a given market. A bilingual focus suggests KORMo-Web-11B is being positioned for real deployment scenarios rather than benchmark demonstrations alone.

A few practical notes from the record:

  • Size: 11B parameters, a dense (non-MoE) model that fits the mid-range tier favored for agent workloads.
  • Modality: vision-language, built for reading and interacting with screens.
  • Languages: Korean and English, with an emphasis on web and desktop GUI tasks.

As with many initial open-weights releases, details such as context length and licensing specifics are still thin, and the model is listed under an "other" license, so teams should review the terms on the Hugging Face repository before building on it.

Sources

  • KORMo-VL/KORMo-Web-11B

    Hugging Face

    Visit
OlderMeituan's LongCat Adds Audio-Driven Avatar VideoMeituan · Image → Video · last weekNewerInternLM's Sage S2.1 Flash targets scientific reasoningInternlm · Reasoning · last week

Get the model

Hugging Face

Specs

Parameters11B
Context window131K tokens
Size22.4 GB
PrecisionBF16
ArchitectureLlavaOnevisionForConditionalGeneration
LicenseOTHER
Downloads1.1K
Likes228

Can you run it?

Runs on a laptop — about 9.1 GB at 4-bit.

  • BF16 (as published)

    32 GB GPU (RTX 5090) · 48 GB Mac

    24.7 GB
  • 8-bit

    16 GB GPU (RTX 4080 / 5070 Ti) · 24 GB Mac

    14.2 GB
  • 4-bit

    12 GB GPU (RTX 3060 / 4070) · 16 GB Mac

    9.1 GB
Your machine
GGUF builds
  • KV cache sized for 8,192 tokens of context; longer prompts need proportionally more.

Based on Hugging Face weights metadata. Assumes an 8K context and ~1 GB runtime overhead; actual needs vary. How we estimate


Modalities

Vision-Language

The Weekly Weights

Every open release that mattered, one email a week.

0 comments

No comments yet. Be the first to weigh in.

More in Vision-Language

All Vision-Language →
SenseNova-Loop-9B
SenseTime/Vision-LanguageRuns on a laptop

SenseTime Releases Looped 9B Multimodal Model

The company's new SenseNova-Loop-9B pairs vision-language capabilities with a recurrent looped-transformer design.

Nov 17, 2026
Hunyuan Hy4-flash
Tencent/Vision-LanguageMulti-GPU server

Tencent releases Hunyuan Hy4-flash multimodal MoE

The lightweight, mixture-of-experts model adds vision and text instruction-following to Tencent's open Hunyuan lineup.

Nov 17, 2026
Inkling 2 Mini
Thinkingmachines/Vision-LanguageConsumer GPU

Thinking Machines releases Inkling 2 Mini, a compact open MoE

The Apache-2.0 multimodal model uses a mixture-of-experts design aimed at running locally rather than in the cloud.

Nov 17, 2026