Instructions to use FINAL-Bench/POCKET-35B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use FINAL-Bench/POCKET-35B-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-35B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf FINAL-Bench/POCKET-35B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf FINAL-Bench/POCKET-35B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf FINAL-Bench/POCKET-35B-GGUF: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 FINAL-Bench/POCKET-35B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf FINAL-Bench/POCKET-35B-GGUF: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 FINAL-Bench/POCKET-35B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf FINAL-Bench/POCKET-35B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/FINAL-Bench/POCKET-35B-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use FINAL-Bench/POCKET-35B-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-35B-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-35B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/FINAL-Bench/POCKET-35B-GGUF:Q4_K_M

Ollama
How to use FINAL-Bench/POCKET-35B-GGUF with Ollama:
```
ollama run hf.co/FINAL-Bench/POCKET-35B-GGUF:Q4_K_M
```

Unsloth Studio

How to use FINAL-Bench/POCKET-35B-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-35B-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-35B-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-35B-GGUF to start chatting

How to use FINAL-Bench/POCKET-35B-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-35B-GGUF: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": "FINAL-Bench/POCKET-35B-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use FINAL-Bench/POCKET-35B-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-35B-GGUF: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 FINAL-Bench/POCKET-35B-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use FINAL-Bench/POCKET-35B-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-35B-GGUF: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 "FINAL-Bench/POCKET-35B-GGUF: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 FINAL-Bench/POCKET-35B-GGUF with Docker Model Runner:
```
docker model run hf.co/FINAL-Bench/POCKET-35B-GGUF:Q4_K_M
```

Lemonade

How to use FINAL-Bench/POCKET-35B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull FINAL-Bench/POCKET-35B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.POCKET-35B-GGUF-Q4_K_M

List all available models

lemonade list

The model is unstable and may enter an infinite loop.

by Salormatik999 - opened 4 days ago

Discussion

Salormatik999

4 days ago

I launched it in agent mode using kilo code. After 30,000 tokens, the model started repeating the same thing and doing the same thing. I use this quantum POCKET-35B-Q4_K_M.gguf. The model is really fast, but not stable.

SeaWolf-AI

FINAL_Bench org 3 days ago

I launched it in agent mode using kilo code. After 30,000 tokens, the model started repeating the same thing and doing the same thing. I use this quantum POCKET-35B-Q4_K_M.gguf. The model is really fast, but not stable.

Hey, thanks a lot for the detailed report — and glad the speed is treating you well! 🙏

That looping after ~30k tokens is a known behavior of this Qwen3.5-family model, and it's almost always a sampling issue in agent mode rather than the weights. Kilo Code tends to run near-greedy, and without a presence penalty these models can fall into repetition on long generations.

Could you try:
temp 0.7, top_p 0.8, top_k 20, min_p 0, presence_penalty 1.5 (avoid temp 0).

That fixes the loop in the vast majority of cases. If it still repeats past 32k, it may be context-length/RoPE scaling — let me know your context window setting and I'll dig in. We'll also add these recommended params to the model card. Thanks again! 🚀

Salormatik999

2 days ago

•

edited 2 days ago

Thank you very much for your attention. It really helped to expand the number of tokens to 65k. llama-server.exe ^ -m POCKET-35B-Q4_K_M.gguf ^ --host 127.0.0.1 --port 5008 ^ -c 64536 ^ -t 8 -tb 11 --mlock ^ --batch-size 1024 --ubatch-size 512 ^ -ctk q8_0 -ctv q8_0 ^ --temp 0.6 --top-p 0.8 --top-k 20 --min-p 0 --presence-penalty 1.5 ^ --jinja -ngl 0

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment