Instructions to use FINAL-Bench/POCKET-EN-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use FINAL-Bench/POCKET-EN-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="FINAL-Bench/POCKET-EN-GGUF",
	filename="POCKET-EN-PC-mix.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 FINAL-Bench/POCKET-EN-GGUF 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 FINAL-Bench/POCKET-EN-GGUF
# Run inference directly in the terminal:
llama cli -hf FINAL-Bench/POCKET-EN-GGUF

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf FINAL-Bench/POCKET-EN-GGUF
# Run inference directly in the terminal:
llama cli -hf FINAL-Bench/POCKET-EN-GGUF

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 FINAL-Bench/POCKET-EN-GGUF
# Run inference directly in the terminal:
./llama-cli -hf FINAL-Bench/POCKET-EN-GGUF

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 FINAL-Bench/POCKET-EN-GGUF
# Run inference directly in the terminal:
./build/bin/llama-cli -hf FINAL-Bench/POCKET-EN-GGUF

Use Docker

docker model run hf.co/FINAL-Bench/POCKET-EN-GGUF

LM Studio
Jan

vLLM

How to use FINAL-Bench/POCKET-EN-GGUF with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/FINAL-Bench/POCKET-EN-GGUF

Ollama
How to use FINAL-Bench/POCKET-EN-GGUF with Ollama:
```
ollama run hf.co/FINAL-Bench/POCKET-EN-GGUF
```

Unsloth Studio

How to use FINAL-Bench/POCKET-EN-GGUF 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 FINAL-Bench/POCKET-EN-GGUF 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 FINAL-Bench/POCKET-EN-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for FINAL-Bench/POCKET-EN-GGUF to start chatting

How to use FINAL-Bench/POCKET-EN-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf FINAL-Bench/POCKET-EN-GGUF

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": "FINAL-Bench/POCKET-EN-GGUF"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use FINAL-Bench/POCKET-EN-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf FINAL-Bench/POCKET-EN-GGUF

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 FINAL-Bench/POCKET-EN-GGUF

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use FINAL-Bench/POCKET-EN-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf FINAL-Bench/POCKET-EN-GGUF

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 "FINAL-Bench/POCKET-EN-GGUF" \
  --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 FINAL-Bench/POCKET-EN-GGUF with Docker Model Runner:
```
docker model run hf.co/FINAL-Bench/POCKET-EN-GGUF
```

Lemonade

How to use FINAL-Bench/POCKET-EN-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull FINAL-Bench/POCKET-EN-GGUF

Run and chat with the model

lemonade run user.POCKET-EN-GGUF-{{QUANT_TAG}}

List all available models

lemonade list

📚 Collections

▶ POCKET Models — this family (on-device, no GPU) Darwin Family · Aether Foundation · VKAE Accelerated · Metacognition Adapters

POCKET-EN-GGUF · English

English-focused pocket build using our MoE mixed-precision trick — keeps the shared layers high-bit so quality holds at 5 GB. Runs on iPhone (PocketPal) and any CPU PC. No fork.

🚀 Try it live, no install → — a 35B model answering on a CPU-only box.

Pick your build →

The POCKET lineup — pick by your device

Repo	File	Size	Runs on	Best for	Korean PPL*
POCKET-35B-GGUF	`Q4_K_M`	21 GB	PC / server (32 GB RAM)	top quality	5.79
POCKET-35B-GGUF	`Q2_K` ⭐	13 GB	mini-PC, no GPU	daily driver	6.49
POCKET-35B-GGUF	`IQ1_M`	8.2 GB	16 GB RAM box	smallest full model	9.69
POCKET-KR-GGUF	`IQ2_M`	5.1 GB	Android 8 GB+	🇰🇷 Korean phone	7.95
POCKET-KR-MLX	2-bit	5.1 GB	🍎 iPhone / iPad / Mac	🇰🇷 Korean, Apple-native	7.95
POCKET-EN-GGUF	`iPhone-mix`	5.3 GB	🍎 iPhone (PocketPal)	🌍 English phone	—
POCKET-EN-GGUF	`PC-mix`	6.8 GB	PC / Android	🌍 English, best quality	—

*Wikipedia-Korean perplexity, lower is better. Q4_K_M = 5.79 baseline. English builds are tuned on English; see each repo.

🍎 Why MLX for Korean but GGUF for English on iPhone? Apple-native MLX only does uniform quantization. Korean survives it (96 experts hold up); English needs our mixed-precision trick, which only GGUF supports — so the English iPhone build ships as a GGUF you run with PocketPal. Honest, not lazy.

Benchmarks — what is measured, what is not

We measure Bonsai on the same machine with the same stock llama.cpp, and we tell you where we lose.

[measured] Generation speed — POCKET wins on both CPU and GPU:

	POCKET-35B IQ1_M	Bonsai-27B Q1_0
CPU generate (Xeon, 16t)	27.0 tok/s	10.1	🟢 2.69×
GPU generate (H100)	197 tok/s	89	🟢 2.22×
GPU prompt (H100)	753	1816	🔴 0.41×
Quality (HellaSwag, 400q)	61.0%	60.0%	⚪ tie (CI overlaps)

[measured on a MacBook M3 Pro, 18 GB] — and on a laptop, POCKET wins every axis, including prompt processing:

	POCKET-35B IQ1_M	Bonsai-27B Q1_0
Metal generate (tg64)	25.4 tok/s	12.8	🟢 1.99×
CPU generate (8 threads)	13.8 tok/s	4.4	🟢 3.13×
Metal prompt (pp128)	240.7 tok/s	73.4	🟢 3.28×
CPU prompt (pp128)	45.5 tok/s	9.6	🟢 4.75×

On a laptop GPU the arithmetic headroom that let Bonsai win prefill on an H100 is gone, so MoE sparsity wins across the board. POCKET-35B-Q2_K runs on the M3 Pro's CPU at 19.5 tok/s — on an 18 GB Mac, run Q2_K on CPU (-ngl 0); its 13 GB exceeds the recommended Metal budget.

[pending — community reports welcome] on-device iPhone and Strix Halo throughput. We publish only what we ran ourselves; help us fill the rest.

The same-size rival Ternary-Bonsai-27B-Q2_0 (7.2 GB) fails to load in upstream llama.cpp — it needs the PrismML fork. POCKET runs on the tools you already have.

Files in this repo

File	Size	Runs on	vs baseline
`POCKET-EN-iPhone-mix.gguf`	5.3 GB	🍎 iPhone (PocketPal)	+57% PPL
`POCKET-EN-PC-mix.gguf`	6.8 GB	PC / Android	+36% PPL (near-full)

Mixed precision = routing experts at 1–2 bit, shared path (attention + shared expert) kept high-bit. English lives across many experts, so protecting the shared path is what saves it.

Quickstart

llama-cli -m POCKET-EN-PC-mix.gguf -p "Explain mixture-of-experts in one line." -ngl 0 -t 8

Lineage — where POCKET comes from

POCKET is quantized from Darwin-36B-Opus, VIDRAFT's flagship — a model bred and evolved over several generations on the Darwin platform (crossbreeding, healing, expert surgery). Darwin-36B-Opus itself traces back to a Qwen3.5-family MoE architecture.

Component	Origin
Starting checkpoint	Darwin-36B-Opus — VIDRAFT, multi-generation Darwin evolution
Base architecture	Qwen3.5-family MoE (256 experts, top-8), unchanged
Quantization (`Q4_K_M`…`IQ1_M`)	stock llama.cpp — no custom format
Runtime	upstream llama.cpp / Apple MLX — unmodified
Expert pruning + domain imatrix (KR/EN builds)	ours (VIDRAFT)

The CPU/GPU speed comes from the sparse-MoE architecture plus ordinary quantization — reproducible with the same base and the same tools. What we add is the Darwin-evolved weights, the honest measurement, the Korean tuning, and the pruning that makes the 5 GB phone builds.

Limitations

The iPhone/Mac speed is not yet measured by us — community reports welcome.
Extreme quants (IQ1_M) hurt Korean ~2.8× more than English; use Q2_K or larger for quality.
English phone builds trade quality for size; the PC build (PC-mix) is much closer to full quality.