Instructions to use esilva/SlopCoder-Mongo-0.5B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use esilva/SlopCoder-Mongo-0.5B with Transformers:

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

pipe = pipeline("text-generation", model="esilva/SlopCoder-Mongo-0.5B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("esilva/SlopCoder-Mongo-0.5B")
model = AutoModelForCausalLM.from_pretrained("esilva/SlopCoder-Mongo-0.5B", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use esilva/SlopCoder-Mongo-0.5B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "esilva/SlopCoder-Mongo-0.5B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "esilva/SlopCoder-Mongo-0.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/esilva/SlopCoder-Mongo-0.5B

SGLang

How to use esilva/SlopCoder-Mongo-0.5B 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 "esilva/SlopCoder-Mongo-0.5B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "esilva/SlopCoder-Mongo-0.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "esilva/SlopCoder-Mongo-0.5B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "esilva/SlopCoder-Mongo-0.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use esilva/SlopCoder-Mongo-0.5B with Docker Model Runner:
```
docker model run hf.co/esilva/SlopCoder-Mongo-0.5B
```

SlopCoder-Mongo-0.5B

File size: 1,972 Bytes

96d8449

# NOTICE

SlopCoder-Mongo models are derivative works of two upstream models. Both licenses apply.

## 1. DeepSeek Coder (DeepSeek License Agreement) — file `LICENSE`

The training data of these models was scored, ranked, filtered and partially labeled by the teacher model
**SlopCoder-Mongo-6.7B-v1**, a QLoRA fine-tune of
[`deepseek-ai/deepseek-coder-6.7b-base`](https://huggingface.co/deepseek-ai/deepseek-coder-6.7b-base).

The DeepSeek License Agreement defines "Derivatives of the Model" to include models created by distillation methods,
including methods based on the generation of synthetic data by the Model. These models are therefore distributed as
Derivatives of the Model under the DeepSeek License Agreement, and:

- a copy of the DeepSeek License Agreement is provided in `LICENSE`;
- **the use-based restrictions of paragraph 5 and Attachment A of the DeepSeek License Agreement apply to these models
  and to any further derivative of them, and are an enforceable condition of their use and distribution**;
- anyone redistributing these models (or derivatives) must pass on these same restrictions and a copy of the license.

No DeepSeek weights are included in these repositories.

Copyright (c) 2023 DeepSeek. DeepSeek Coder code is released under the MIT License.

## 2. Qwen2.5-Coder (Apache License 2.0) — file `LICENSE-APACHE-2.0`

The weights were initialized from `Qwen/Qwen2.5-Coder-0.5B` or `Qwen/Qwen2.5-Coder-1.5B` (Alibaba Cloud, Qwen team),
licensed under the Apache License, Version 2.0.

Modifications: LoRA fine-tuning (r=64, all attention and MLP projections) merged into the bf16 weights; for the `-ONNX`
repositories, export to ONNX Runtime GenAI and INT4/INT8 weight quantization.

## 3. Training data

100% synthetic data generated by the SlopStudio.MongoAI pipelines (fictitious schemas and values, a MongoDB operator
catalog, and examples from the Slop Studio Console documentation). No customer data, no scraped web content.