Instructions to use p-p-n/Huvm with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use p-p-n/Huvm with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf p-p-n/Huvm:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf p-p-n/Huvm:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf p-p-n/Huvm:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf p-p-n/Huvm:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf p-p-n/Huvm:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf p-p-n/Huvm:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf p-p-n/Huvm:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf p-p-n/Huvm:Q4_K_M

Use Docker

docker model run hf.co/p-p-n/Huvm:Q4_K_M

LM Studio
Jan

vLLM

How to use p-p-n/Huvm with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/p-p-n/Huvm:Q4_K_M

Ollama
How to use p-p-n/Huvm with Ollama:
```
ollama run hf.co/p-p-n/Huvm:Q4_K_M
```
Unsloth Desktop

How to use p-p-n/Huvm with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf p-p-n/Huvm:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "p-p-n/Huvm:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use p-p-n/Huvm with Docker Model Runner:
```
docker model run hf.co/p-p-n/Huvm:Q4_K_M
```

Lemonade

How to use p-p-n/Huvm with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull p-p-n/Huvm:Q4_K_M

Run and chat with the model

lemonade run user.Huvm-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use p-p-n/Huvm with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf p-p-n/Huvm:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default p-p-n/Huvm:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use p-p-n/Huvm with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf p-p-n/Huvm:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "p-p-n/Huvm:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Huvm

Huvm is a compact, instruction-tuned language model fine-tuned from Qwen/Qwen2.5-3B-Instruct using LoRA + DPO. It is designed to be fast, precise, and carry a distinct personality — a blend of 50% Grok (direct, lightly sarcastic, no-nonsense) and 50% Claude (articulate, deep, careful).

The name comes from "humm" (the sound of thinking) plus the letter V for speed and truth.

Model Description

Base model: Qwen/Qwen2.5-3B-Instruct
Fine-tuning: LoRA (SFT) on 122 curated examples, then DPO (29 preference pairs)
Training: CPU-only (AMD Ryzen 5, 16 GB RAM)
Formats: Hugging Face Transformers, GGUF (f16, q4_k_m)
Reference date: 3 September 2026 (no real-time access)

Intended Use

Huvm is built for local, offline use — desktop CPU or mobile via GGUF. It handles:

💻 Code — explains the why
➗ Math — shows the reasoning
💬 Multilingual chat — replies in the user's language
🎨 Creative writing — avoids generic fluff

It is not intended for high-stakes, medical, legal, or safety-critical applications, nor for real-time news, statistics, or link retrieval.

Personality & Tone

Trait	Behavior
Directness	Gets to the point, minimal filler
Sarcasm	Light and playful, never insulting
Honesty	Refuses to invent facts, links, or stats
Language	Replies in the user's language (pt/en/es)

Bias, Risks & Limitations

Important limitations — read before use.

Hallucination risk on real-time facts. Despite refusal training and DPO, Huvm can still produce generic or made-up answers when asked about breaking news, current statistics, population numbers, or specific URLs/links. It often refuses, but not always. Do not rely on it for live facts; verify with primary sources.
Small model. 3B parameters means limited capacity and occasional repetition or shallow answers on complex topics.
Fixed knowledge cutoff. Reference date is 2026-09-03; it has no internet access and cannot browse.
Training data. Identity and behavior are heavily reinforced, but coverage is narrow; outside the training distribution the base model's priors dominate.
No additional safety tuning. Standard base-model safety applies; use with judgement.

Training Details

SFT: 122 curated examples (identity, multilingual, code, math, creativity, anti-hallucination flows) — 3 epochs, lr 1e-4, LoRA r=32/alpha=64.
DPO: 29 chosen/rejected pairs targeting hallucinated news, stats, and links — 5 epochs, lr 5e-6, beta 0.1.
Hardware: AMD Ryzen 5 5600GT, 16 GB RAM, CPU-only.

Quantization

File	Size	Note
`huvm-q4_k_m.gguf`	~1.93 GB	Recommended — balanced quality/space

Run with llama.cpp:

llama-cli -m huvm-q4_k_m.gguf \
  -p "<|im_start|>system\nEu sou o Huvm...<|im_end|>\n<|im_start|>user\nQuem e voce?<|im_end|>\n<|im_start|>assistant\n" \
  -n 200 --temp 0.0

Usage (Transformers)

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("your-username/huvm")
tokenizer = AutoTokenizer.from_pretrained("your-username/huvm")

SYSTEM = "Eu sou o Huvm, um assistente de IA pessoal, rapido, preciso e com personalidade. Metade Grok (direto, sarcasmo leve e bem-humorado), metade Claude (articulado, profundo, cuidadoso). Data de referencia: 3 de setembro de 2026. Respondo no idioma do usuario. No codigo eu explico o por que; na matematica mostro o raciocinio; na criatividade fujo do generico. NUNCA invento fatos, fontes, links, citacoes, estatisticas ou noticias. Respondo SEMPRE diretamente, sem bloco de raciocinio, seja objetivo."

messages = [
    {"role": "system", "content": SYSTEM},
    {"role": "user", "content": "Faca uma funcao em Python que inverte uma string"},
]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt")
out = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(out[0], skip_special_tokens=True))

Acknowledgements

Built on Qwen 2.5 (Alibaba), TRL (Hugging Face), and llama.cpp.

Downloads last month: 40

GGUF

Model size

3B params

Architecture

qwen2

Hardware compatibility

4-bit

Model tree for p-p-n/Huvm

Base model

Qwen/Qwen2.5-3B

Finetuned

Qwen/Qwen2.5-3B-Instruct

Quantized

(287)

this model