Instructions to use TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b",
	filename="qwen3-vl-8b-instruct.Sumtables-BF16-mmproj.gguf",
)

llm.create_chat_completion(
	messages = "\"Меня зовут Вольфганг и я живу в Берлине\""
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b: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 TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b: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 TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b:Q4_K_M

Use Docker

docker model run hf.co/TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b:Q4_K_M

LM Studio
Jan
Ollama
How to use TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b with Ollama:
```
ollama run hf.co/TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b:Q4_K_M
```

Unsloth Studio

How to use TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b 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 TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b 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 TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b to start chatting

How to use TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b: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": "TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b: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 TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b with Docker Model Runner:
```
docker model run hf.co/TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b:Q4_K_M
```

Lemonade

How to use TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b-Q4_K_M

List all available models

lemonade list

Hang tight. 6/1/2026: Training is done and this is a much more capable model than Gemma4-E4B. After the training it had very deep understanding, but had a issue where the EoF was blurred because of a training setting (minor issue). This caused a couple of similar signs to be confused, but very strong understanding overall. I most likely will retrain it in the next day or two, and while that is being fixed here are the Q4, Q5, Q6 and Q8 for this version.

Yes.. Gemma4-E4B will also benifit from the EoF blur that was caught, and i will lower the LR to deepen its understanding which will make it far more capable. 🍻

Downloads last month: 6

GGUF

Model size

8B params

Architecture

qwen3vl

Hardware compatibility

4-bit

5-bit

6-bit

8-bit

Model tree for TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b

Base model

Qwen/Qwen3-VL-8B-Instruct

Quantized

(82)

this model

TRACCERR
/

Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b

Model tree for TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b

Dataset used to train TRACCERR/Qwen3-VL-8B-Instruct-Sumtablets-V1-bnb-4b