Instructions to use PromethicLabs/Emberon-1.2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use PromethicLabs/Emberon-1.2B with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="PromethicLabs/Emberon-1.2B",
	filename="Emberon-1.2B-F16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use PromethicLabs/Emberon-1.2B 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 PromethicLabs/Emberon-1.2B:F16
# Run inference directly in the terminal:
llama cli -hf PromethicLabs/Emberon-1.2B:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf PromethicLabs/Emberon-1.2B:F16
# Run inference directly in the terminal:
llama cli -hf PromethicLabs/Emberon-1.2B:F16

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 PromethicLabs/Emberon-1.2B:F16
# Run inference directly in the terminal:
./llama-cli -hf PromethicLabs/Emberon-1.2B:F16

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 PromethicLabs/Emberon-1.2B:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf PromethicLabs/Emberon-1.2B:F16

Use Docker

docker model run hf.co/PromethicLabs/Emberon-1.2B:F16

LM Studio
Jan

vLLM

How to use PromethicLabs/Emberon-1.2B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "PromethicLabs/Emberon-1.2B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "PromethicLabs/Emberon-1.2B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/PromethicLabs/Emberon-1.2B:F16

Ollama
How to use PromethicLabs/Emberon-1.2B with Ollama:
```
ollama run hf.co/PromethicLabs/Emberon-1.2B:F16
```

Unsloth Studio

How to use PromethicLabs/Emberon-1.2B with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for PromethicLabs/Emberon-1.2B to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for PromethicLabs/Emberon-1.2B to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for PromethicLabs/Emberon-1.2B to start chatting

How to use PromethicLabs/Emberon-1.2B with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf PromethicLabs/Emberon-1.2B:F16

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "PromethicLabs/Emberon-1.2B:F16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use PromethicLabs/Emberon-1.2B with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf PromethicLabs/Emberon-1.2B:F16

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 PromethicLabs/Emberon-1.2B:F16

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use PromethicLabs/Emberon-1.2B with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf PromethicLabs/Emberon-1.2B:F16

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 "PromethicLabs/Emberon-1.2B:F16" \
  --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"

Docker Model Runner
How to use PromethicLabs/Emberon-1.2B with Docker Model Runner:
```
docker model run hf.co/PromethicLabs/Emberon-1.2B:F16
```

Lemonade

How to use PromethicLabs/Emberon-1.2B with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull PromethicLabs/Emberon-1.2B:F16

Run and chat with the model

lemonade run user.Emberon-1.2B-F16

List all available models

lemonade list

Promethic commited on 19 days ago

Commit

c812f29

verified ·

1 Parent(s): 3a25c9d

Add real-world held-out eval (n=79 unseen live-usage dictations): 0% hallucination, 9/9 suppression

Browse files

Files changed (1) hide show

README.md +16 -0

README.md CHANGED Viewed

@@ -149,6 +149,22 @@ or turning *"open paren"* into `(` — so the output is *supposed* to diverge fr
 The real-dictation subset performs **at least as well as** synthetic — evidence the behavior is not an artifact
 of the synthetic training distribution.
 ### Performance (Apple Silicon, Metal, as the app runs it)
 | | Q4_K_M |

 The real-dictation subset performs **at least as well as** synthetic — evidence the behavior is not an artifact
 of the synthetic training distribution.
+### Real-world held-out (unseen live usage)
+As an out-of-distribution check, we evaluated on **79 real dictations captured from live app usage** — strictly
+leakage-filtered against *all* training/eval data, deduped, and much longer than the eval set (median **34 words**;
+these are real, messy, agentic prompts):
+| Metric | Result |
+|---|---|
+| **Content-addition / hallucination** | **0.00%** (0/79) |
+| Mean novelty (lower = more faithful) | 0.009 |
+| **Suppression** (answer-tempting subset) | **9/9 = 100%** |
+Zero hallucinations across 79 genuinely-unseen, long real-world prompts, and it answered none of the real spoken
+questions. *(Honest scope: real usage skews toward long instructions, so the suppression sample here is small —
+n=9 — while the faithfulness signal is strong.)*
 ### Performance (Apple Silicon, Metal, as the app runs it)
 | | Q4_K_M |