turbolego/ qwen3.5-0.8b-wcag-gauntlet-lora-q4_k_m

15 Downloads Updated 2 days ago

ollama run turbolego/qwen3.5-0.8b-wcag-gauntlet-lora-q4_k_m

curl http://localhost:11434/api/chat \
  -d '{
    "model": "turbolego/qwen3.5-0.8b-wcag-gauntlet-lora-q4_k_m",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='turbolego/qwen3.5-0.8b-wcag-gauntlet-lora-q4_k_m',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'turbolego/qwen3.5-0.8b-wcag-gauntlet-lora-q4_k_m',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Models

Name

1 model

Size / Usage

Context

Input

qwen3.5-0.8b-wcag-gauntlet-lora-q4_k_m:latest

529MB · 256K context window · Text · 2 days ago

qwen3.5-0.8b-wcag-gauntlet-lora-q4_k_m:latest

529MB

256K

Text

Readme

Qwen3.5 0.8B WCAG Gauntlet LoRA — Q4_K_M

This is the LoRA-adapted Qwen3.5-0.8B model used to evaluate how the LoRA changes affect performance on the AI-WCAG-Gauntlet benchmark. The benchmark assesses an AI agent’s ability to build an HTML5 showcase that meets WCAG 2.2 Level AAA requirements, including iterative validation and accessibility testing.

Run

ollama run turbolego/qwen3.5-0.8b-wcag-gauntlet-lora-q4_k_m

Comparison

Compare this model with the paired unadapted baseline, turbolego/qwen3.5-0.8b-base-q4_k_m. Both models use Q4_K_M quantization and the same Ollama prompt template.

For a fair A/B comparison, use the same benchmark instructions, tasks, tools, runtime settings, and evaluation process for both models. Keep the model as the only intentional variable, then compare the benchmark results, including the Agent Performance Score, validation and accessibility findings, HTML5 tag coverage, and iterations required.

Base model reference

# Qwen3.5 0.8B WCAG Gauntlet LoRA — Q4_K_M

This is the LoRA-adapted Qwen3.5-0.8B model used to evaluate how the LoRA changes affect performance on the [AI-WCAG-Gauntlet benchmark](https://github.com/turbolego/AI-WCAG-Gauntlet). The benchmark assesses an AI agent's ability to build an HTML5 showcase that meets WCAG 2.2 Level AAA requirements, including iterative validation and accessibility testing.

## Run

```bash
ollama run turbolego/qwen3.5-0.8b-wcag-gauntlet-lora-q4_k_m
```

## Comparison

Compare this model with the paired unadapted baseline, [`turbolego/qwen3.5-0.8b-base-q4_k_m`](https://ollama.com/turbolego/qwen3.5-0.8b-base-q4_k_m). Both models use Q4_K_M quantization and the same Ollama prompt template.

For a fair A/B comparison, use the same benchmark instructions, tasks, tools, runtime settings, and evaluation process for both models. Keep the model as the only intentional variable, then compare the benchmark results, including the Agent Performance Score, validation and accessibility findings, HTML5 tag coverage, and iterations required.

## Base model reference

- [Unsloth Qwen3.5-0.8B GGUF](https://huggingface.co/unsloth/Qwen3.5-0.8B-GGUF)
- [AI-WCAG-Gauntlet benchmark](https://github.com/turbolego/AI-WCAG-Gauntlet)

Paste, drop or click to upload images (.png, .jpeg, .jpg, .svg, .gif)