KORMo-Web-4B is a 4B vision-language agent that completes real browser tasks from screenshots alone. Give it a request in Korean or English; it plans, clicks, types and moves between sites and tabs until the task is done. In the same live-web evaluation, it beats every open-weight web agent we compared, by +9.6 points in English and +22.1 points in Korean.
On Online-Mind2Web it reaches 42.3% in English and 42.3% in Korean, which is +22.3 / +28.8 points over the base Qwen3.5-4B.
All models below ran the same 612 live-website tasks (300 English, 312 Korean), each with its own prompt and action format, and were scored by the same LLM judge.
| Model | Size | Online-Mind2Web EN | Online-Mind2Web KO |
|---|---|---|---|
| Qwen3.5-4B (base, no fine-tuning) | 4B | 20.0 | 13.5 |
| Fara-1.5-4B | 4B | 20.7 | 7.7 |
| UI-TARS-1.5-7B | 7B | 22.3 | 20.2 |
| OpenWebRL-4B | 4B | 32.7 | 15.4 |
| KORMo-Web-4B (ours) | 4B | 42.3 | 42.3 |
| KORMo-Web-9B (ours) | 9B | 51.0 | 48.1 |
Success rate (%). The denominator is always the full task count, so crashed or empty runs count as failures.
KORMo-Web-4B outperforms open web agents of similar size in both English and Korean.
This repository holds the 4-bit MLX weights of KORMo-Web-4B for Apple Silicon Macs. Serve it with vllm-mlx (0.6.0 or later), which exposes the same OpenAI-compatible API as vLLM.
1. Install and download
pip install -U "vllm-mlx>=0.6.0" openai playwright huggingface_hub
playwright install chromium
huggingface-cli download KORMo-VL/KORMo-Web-4B-mlx-4bit --local-dir KORMo-Web-4B-mlx-4bit
# system prompt, tool schema and reference harness (shared with the full-precision repo)
huggingface-cli download KORMo-VL/KORMo-Web-4B system_prompt.md tools.json HARNESS.md examples/agent_playwright.py \
--local-dir KORMo-Web-4B-mlx-4bit
2. Start the model server
vllm-mlx serve ./KORMo-Web-4B-mlx-4bit --served-model-name KORMo-Web-4B --mllm --port 8000
Do not enable --enable-auto-tool-choice or --reasoning-parser: the harness parses the raw <think> and <tool_call> text from the message content.
3. Run a browser task
python KORMo-Web-4B-mlx-4bit/examples/agent_playwright.py --model KORMo-Web-4B \
--endpoint http://localhost:8000/v1 \
--task "๋ค๋ฐฉ์์ ์ํ์ญ ๊ทผ์ฒ ๋ณด์ฆ๊ธ ์ฒ๋ง์ ์ดํ๋ก ์ค์ ํ๋ค ์๋ฃธ 3๊ฐ์ ๋ ์ฐพ์์ ์ดํด๋ณด๊ณ ๋คํ๋ฉด ์ฐพ์ ๋ด์ฉ์ ์นด์ด์คํธ ๋ฉ์ผ๋ก mlplab@kaist.ac.kr ํํ
์ ๋ฆฌํด์ ๋ณด๋ด์ค"
The agent opens Chromium, prints each action as it goes, and finishes with the model's answer.
Send one screenshot with the request and get back one action:
import base64, json
from openai import OpenAI
SYSTEM = open("KORMo-Web-4B-mlx-4bit/system_prompt.md").read().strip()
TOOLS = json.load(open("KORMo-Web-4B-mlx-4bit/tools.json"))
shot = base64.b64encode(open("screen.png", "rb").read()).decode()
obs = ("<observation>\nscreen size: 1280 x 720\nCurrent tab: 0\nAvailable tabs:\n"
"- Tab 0 (active): https://www.google.com/\n\nurl: https://www.google.com/\nscreenshot:\n")
messages = [
{"role": "system", "content": SYSTEM},
{"role": "user", "content": [
{"type": "text", "text": "<user_request>\n๊ตฌ๊ธ์์ KAIST ๋ฅผ ๊ฒ์ํด์ ๊ณต์ ํํ์ด์ง๋ฅผ ์ด์ด์ค.\n</user_request>\n" + obs},
{"type": "image_url", "image_url": {"url": f"data:image/png;base64,{shot}"}},
{"type": "text", "text": "\n</observation>"},
]},
]
client = OpenAI(base_url="http://localhost:8000/v1", api_key="EMPTY") # vllm-mlx server
out = client.chat.completions.create(
model="KORMo-Web-4B", messages=messages, tools=TOOLS,
temperature=0.6, top_p=0.95, max_tokens=16384,
extra_body={"top_k": 20, "chat_template_kwargs": {"enable_thinking": True}},
)
print(out.choices[0].message.content)
The output looks like this (illustrative):
<think>
I can see the Google home page with an empty search box in the center. I have not searched yet.
My plan: 1. Click the search box. 2. Type "KAIST". 3. Press Enter and open the official site.
</think>
<tool_call>
<function=click>
<parameter=x>
500
</parameter>
<parameter=y>
427
</parameter>
</function>
</tool_call>
Coordinates in the tool call are normalized to 0โ1000. Convert them to pixels with x / 1000 ร width and y / 1000 ร height.
user <user_request> task </user_request> + observation (tabs, url, screenshot)
assistant <think> what I see ยท what is done ยท plan </think> <tool_call> one action </tool_call>
tool action result + new observation (newest screenshot only)
โฆ repeat until finished(answer)
finished, and others (tools.json).generation_config.json.The exact message format, parsing rules and feedback strings are in HARNESS.md. Follow it if you build your own harness.
| Tasks | Online-Mind2Web: 300 English, 312 Korean |
| Judge | GPT-6 |
| File | Description |
|---|---|
chat_template.jinja |
Chat template (Qwen3.5 XML tool calls, thinking on) |
system_prompt.md ยท tools.json |
System prompt and 22-tool schema, verbatim from training |
HARNESS.md |
Full harness specification |
examples/agent_playwright.py |
Reference browser agent (Playwright + OpenAI-compatible API). It does not execute upload. |
config.json ยท generation_config.json ยท processor_config.json ยท tokenizer*.json |
Model, sampling and processor configs |
KORMo-Web-4B๋ ์ฌ๋์ฒ๋ผ ์น ํ๋ฉด์ ๋ณด๊ณ ํด๋ฆญยท์ ๋ ฅยทํ์์ ์ํํ๋ 4B ๊ท๋ชจ์ ํ์ ์๊ฐ ๊ธฐ๋ฐ ์น ์์ด์ ํธ์ ๋๋ค.
Built on Qwen3.5-4B. Evaluated on Online-Mind2Web and its Korean counterpart.
KORMo-VL/KORMo-Web-4B