---
type: "post"
id: "inclusionai-releases-lbs-27b-a-qwen3-8-27b-214d7f"
slug: "inclusionai-releases-lbs-27b-a-qwen3-8-27b-214d7f"
title: "inclusionAI releases LBS-27B, a Qwen3.8-27B vision-language agent trained only on synthetic GUI-grounding data"
description: "inclusionAI has released LBS-27B, or Learning to Look Beyond Screens, on Hugging Face under the Apache-2.0 licence."
retrieval_nugget: "inclusionAI has released LBS-27B, or Learning to Look Beyond Screens, on Hugging Face under the Apache-2.0 licence."
published_at: "2026-10-27"
updated_at: "2026-10-27"
record_date: "2026-10-27"
date_kind: "published_at"
topics: ["computer-use","models","open-models"]
entities: ["inclusionAI","LBS-27B","Qwen3.8-27B","UI-Venus","Hugging Face"]
related_patterns: []
novelty_label: "new_fact"
what_changed: "InclusionAI adds LBS-27B, a Qwen3.8-27B vision-language agent trained only on synthetic GUI-grounding data; the prior covered its separate Ling-3.1-flash release."
prior_refs: ["memory:05d8646c-e2ac-8615-ac8e-6805d94503a7"]
source_url: "https://huggingface.co/inclusionAI/LBS-27B"
source_urls: ["https://huggingface.co/inclusionAI/LBS-27B"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:inclusionai-releases-lbs-27b-a-qwen3-8-27b-214d7f"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/inclusionai-releases-lbs-27b-a-qwen3-8-27b-214d7f/","markdown":"https://newruntime.com/posts/inclusionai-releases-lbs-27b-a-qwen3-8-27b-214d7f.md","json":"https://newruntime.com/posts/inclusionai-releases-lbs-27b-a-qwen3-8-27b-214d7f.json"}
---

# inclusionAI releases LBS-27B, a Qwen3.8-27B vision-language agent trained only on synthetic GUI-grounding data

## Retrieval answer

inclusionAI has released LBS-27B, or Learning to Look Beyond Screens, on Hugging Face under the Apache-2.0 licence.

inclusionAI has released LBS-27B, or Learning to Look Beyond Screens, on Hugging Face under the Apache-2.0 licence. The 27B vision-language agent model is built on Qwen3.8-27B. Its code is available in the LBS branch of the UI-Venus repository.

The model is designed to select and zoom image regions while answering. The model card specifies up to two crops per turn and three crop calls per episode. inclusionAI says training used about 10,000 synthetic GUI-grounding SFT examples and 80 GRPO steps, with no general-domain VQA data.

With agentic zoom-in, the card reports 81.15% on ScreenSpot-Pro, 84.57% on OS-World-G (Refine), 59.67% on PerceptionBench, 89.13% on HR-Bench 4K, and 87.75% on HR-Bench 8K. PerceptionBench used a self-hosted DeepSeek judge. The source does not establish that these results are directly comparable with other models; it explicitly says chart comparisons are not controlled.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Computer Use](https://newruntime.com/topics/computer-use.md): Explore the computer-use topic hub.
- topic hub: [Models](https://newruntime.com/topics/models.md): Explore the models topic hub.
- topic hub: [Open Models](https://newruntime.com/topics/open-models.md): Explore the open-models topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)