Instructions to use Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated",
	filename="Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated 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 Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated: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 Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated: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 Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Q4_K_M

Use Docker

docker model run hf.co/Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Q4_K_M

LM Studio
Jan
Ollama
How to use Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated with Ollama:
```
ollama run hf.co/Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Q4_K_M
```

Unsloth Studio

How to use Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated 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 Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated 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 Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated to start chatting

How to use Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Q4_K_M

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": "Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated: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 Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated: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 "Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated: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"

Docker Model Runner
How to use Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated with Docker Model Runner:
```
docker model run hf.co/Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Q4_K_M
```

Lemonade

How to use Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Q4_K_M

Run and chat with the model

lemonade run user.Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated-Q4_K_M

List all available models

lemonade list

Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated / QUALITY.md

Chungulus

Add files using upload-large-folder tool

93cce49 verified 16 days ago

preview code

Raw

History Blame Contribute Delete

1.93 kB

Agents-A1 imatrix quality evaluation

This is a local sanity benchmark, not a substitute for MMLU, HellaSwag, or human review. No files were uploaded.

Setup

Accuracy benchmark: eval/quality_multiple_choice.bin
KL holdout text: eval/kl_holdout.txt
KL context/chunks: 128 ctx, 2 chunks
Backend: native llama.cpp, CPU-only (-dev none -ngl 0)

Accuracy Retention

F16 baseline accuracy: 89.5833%

Variant	Accuracy	Retention vs F16	Random chance	Status
A1-Q2_K-imatrix	87.5000%	97.6745%	25.0000%	passed
A1-IQ3_M-imatrix	89.5833%	100.0000%	25.0000%	passed
A1-Q3_K_M-imatrix	89.5833%	100.0000%	25.0000%	passed
A1-Q4_K_M-imatrix	87.5000%	97.6745%	25.0000%	passed
A1-IQ4_XS-imatrix	87.5000%	97.6745%	25.0000%	passed

KL Divergence

Baseline logits file: /Users/oz/Documents/AgentsA1 Quantization/a1-local-quant-pipeline/benchmark_results/quality/baseline_ctx128_chunks2.kld Baseline PPL on KL holdout: 13.0194

Variant	Mean KLD	PPL ratio	PPL delta	Same top p	RMS delta p	Status
A1-Q2_K-imatrix	0.128242	1.158739	2.0667	81.7460%	8.8070%	passed
A1-IQ3_M-imatrix	0.045312	1.001578	0.0205	92.0630%	5.0240%	passed
A1-Q3_K_M-imatrix	0.046980	0.984125	-0.2067	88.0950%	5.2290%	passed
A1-Q4_K_M-imatrix	0.015182	0.986453	-0.1764	93.6510%	2.7470%	passed
A1-IQ4_XS-imatrix	0.020185	0.998199	-0.0234	92.0630%	3.1110%	passed

Notes

Accuracy is a local multiple-choice benchmark generated in eval/quality_multiple_choice.json.
Retention is quant accuracy divided by F16 baseline accuracy on the same local benchmark.
KL divergence compares quant logit distributions to the F16 no-MTP GGUF baseline on eval/kl_holdout.txt.
Smaller KLD is better; higher same-top-p is better.