@@ -92,11 +92,16 @@ git clone https://github.com/MigoXLab/webqa-agent.git
9292cd webqa-agent
9393```
9494
95- Install Python >= 3.10 and run the following commands :
95+ 1 . Recommended: Install dependencies with [ uv ] ( https://github.com/astral-sh/uv ) (Python>=3.11) :
9696
9797``` bash
98- pip install -r requirements.txt
99- playwright install
98+ uv sync
99+ ```
100+
101+ 2 . Install Chromium browser:
102+
103+ ``` bash
104+ uv run playwright install chromium
100105```
101106
102107Performance Analysis - Lighthouse (Optional)
@@ -124,9 +129,25 @@ nuclei -version # Verify successful installation
124129After configuring ` config/config.yaml ` (refer to "Usage > Test Configuration"), run:
125130
126131``` bash
127- python webqa-agent.py
132+ uv run python webqa-agent.py
128133```
129134
135+ ### 🖥️ Visual Web Interface
136+
137+ WebQA Agent provides a user-friendly visual interface powered by Gradio.
138+
139+ ``` bash
140+ # Install Gradio
141+ uv add " gradio>5.44.0"
142+ # Launch the Web UI
143+ uv run python app.py
144+
145+ # Launch with Chinese interface
146+ GRADIO_LANGUAGE=zh-CN uv run python app.py
147+ ```
148+
149+ Access the interface at http://localhost:7860 .
150+
130151## Usage
131152
132153### Test Configuration
@@ -192,14 +213,14 @@ UX (User Experience) testing focuses on usability and user-friendliness. Uses mu
192213
193214# ## 🧠 Recommended Models
194215
195- Based on our testing, these models work well with WebQA Agent :
216+ The following models have been adapted and verified, and are recommended :
196217
197- | Model | Key Strengths | Notes |
198- |-----------------------------------|-----------------------------|---------------------------------|
199- | **gpt-4.1-2025-04-14** ⭐ | High accuracy & reliability | **Best choice** |
200- | **gpt-4.1-mini-2025-04-14** | Cost-effective | ** Economical and practical** |
201- | **qwen3-vl-235b-a22b-instruct** | Open-source, GPT-4.1 level | **Best for on-premise** |
202- | **doubao-seed-1-6-vision-250815** | Vision capabilities | **Excellent web understanding** |
218+ | Model | Recommendation |
219+ |-----------------------------------|-----------------------------|
220+ | **gpt-4.1-2025-04-14** | High accuracy and reliability |
221+ | **gpt-4.1-mini-2025-04-14** | Economical and practical |
222+ | **qwen3-vl-235b-a22b-instruct** | Open-source model, preferred for on-premise |
223+ | **doubao-seed-1-6-vision-250815** | Good web understanding, supports visual recognition |
203224
204225
205226# ## View Results
0 commit comments