Instructions to use caikybaldo999/webcoder-pro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use caikybaldo999/webcoder-pro with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="caikybaldo999/webcoder-pro")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("caikybaldo999/webcoder-pro")
model = AutoModelForCausalLM.from_pretrained("caikybaldo999/webcoder-pro", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use caikybaldo999/webcoder-pro with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "caikybaldo999/webcoder-pro"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "caikybaldo999/webcoder-pro",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/caikybaldo999/webcoder-pro

SGLang

How to use caikybaldo999/webcoder-pro with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "caikybaldo999/webcoder-pro" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "caikybaldo999/webcoder-pro",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "caikybaldo999/webcoder-pro" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "caikybaldo999/webcoder-pro",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use caikybaldo999/webcoder-pro with Docker Model Runner:
```
docker model run hf.co/caikybaldo999/webcoder-pro
```

webcoder-pro / README.md

caikybaldo999

Upload WebCoder-100M — step 8000

7070f0a verified 1 day ago

preview code

Raw

History Blame Contribute Delete

1.65 kB

metadata

library_name: transformers
pipeline_tag: text-generation
language:
  - code
  - pt
  - en
tags:
  - html
  - css
  - javascript
  - code-generation
  - from-scratch
datasets:
  - bigcode/the-stack-smol-xl
license: other

WebCoder-100M

WebCoder-100M é um pequeno modelo decoder-only treinado do zero para geração e continuação de HTML, CSS e JavaScript.

Arquitetura

Parâmetros: 100,000,512 (100.000512M)
Layers: 11
Hidden size: 768
Attention heads: 12
Contexto: 1024 tokens
Vocabulário: 27,664
Precisão de treino: BF16

Dataset

Treinado a partir dos subconjuntos HTML, CSS e JavaScript de bigcode/the-stack-smol-xl, com limpeza, deduplicação e filtro opcional por metadados de licença.

O código-fonte original do dataset pode possuir licenças variadas. Consulte o dataset e a origem de cada amostra antes de usar os pesos em cenários que exijam revisão jurídica específica.

Formato de prompt

<|bos|><|prompt|>
Crie uma landing page responsiva.
<|html|>

Também existem os tokens <|css|> e <|javascript|>.

Métricas desta execução

Step final: 8000
Eval loss: 1.2921
Perplexidade: 3.6405
Tempo de treino: 47.43 min

Limitações

Este é um modelo de aproximadamente 100M parâmetros e treinamento curto. Pode produzir HTML/CSS/JS inválido, inseguro, incompleto ou repetitivo. Revise o código antes de executar ou publicar.

Uso com Transformers

from transformers import AutoTokenizer, AutoModelForCausalLM

tok = AutoTokenizer.from_pretrained("caikybaldo999/webcoder-pro")
model = AutoModelForCausalLM.from_pretrained("caikybaldo999/webcoder-pro")