Instructions to use Jfish123/Qwen3.8-27B-16GB-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 Jfish123/Qwen3.8-27B-16GB-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 Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS
# Run inference directly in the terminal:
llama cli -hf Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS
# Run inference directly in the terminal:
llama cli -hf Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS
# Run inference directly in the terminal:
./build/bin/llama cli -hf Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS

Use Docker

docker model run hf.co/Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS

Llama new
LM Studio
Jan

vLLM

How to use Jfish123/Qwen3.8-27B-16GB-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Jfish123/Qwen3.8-27B-16GB-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": "Jfish123/Qwen3.8-27B-16GB-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS

Ollama
How to use Jfish123/Qwen3.8-27B-16GB-GGUF with Ollama:
```
ollama run hf.co/Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS
```
Unsloth Desktop

How to use Jfish123/Qwen3.8-27B-16GB-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Jfish123/Qwen3.8-27B-16GB-GGUF with Docker Model Runner:
```
docker model run hf.co/Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS
```

Lemonade

How to use Jfish123/Qwen3.8-27B-16GB-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS

Run and chat with the model

lemonade run user.Qwen3.8-27B-16GB-GGUF-UD-IQ3_XXS

List all available models

lemonade list

Hermes Agent

How to use Jfish123/Qwen3.8-27B-16GB-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 Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS

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 Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Jfish123/Qwen3.8-27B-16GB-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS

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 "Jfish123/Qwen3.8-27B-16GB-GGUF:UD-IQ3_XXS" \
  --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"

Qwen3.8-27B for 16 GB GPUs: 125k context, fixed chat template

Qwen3.8-27B squeezed onto a single 16 GB card (tested on an RTX 5080) with 125,000 tokens of context, 100% on the GPU, plus two chat-template fixes so it works in Claude Code and OpenAI Codex through Ollama.

Run it with Ollama:

ollama run awakenedsniper78/qwen3.8-coder

What this is

Weights: unsloth/Qwen3.8-27B-GGUF UD-IQ3_XXS, byte-for-byte unchanged (all 866 tensors verified). Only the embedded tokenizer.chat_template was edited.
Text only: the vision projector is left out; its ~0.9 GB of VRAM goes to context instead.
Thinking on by default, sampling as Qwen recommends for thinking mode (temperature 1.0, top_p 0.95, top_k 20).

Measured on an RTX 5080 (16 GB)


Context	125,000 tokens
GPU offload	66/66 layers (100% GPU)
VRAM	~10.2 GB weights + 2.1 GB KV cache (q4_0) + 0.7 GB compute
Generation speed	~52 tokens/s
Prompt processing	~1,000–2,000 tokens/s
Load time	~5 s from NVMe

Qwen3.8's hybrid design makes long context cheap: only 16 of its 64 layers keep a KV cache (the rest are Gated DeltaNet), so 125k tokens of q4_0 cache is about 2.1 GB.

Required Ollama server settings

The 125k context only fits with a quantized KV cache. Set these environment variables for the Ollama server, then restart Ollama:

OLLAMA_FLASH_ATTENTION=1
OLLAMA_KV_CACHE_TYPE=q4_0
OLLAMA_GPU_OVERHEAD=536870912

OLLAMA_KV_CACHE_TYPE=q4_0: at the default (f16) or q8_0, 125k context doesn't fit next to the weights and layers spill to the CPU.
OLLAMA_GPU_OVERHEAD (512 MB): with the default of 0, Ollama packs VRAM completely full, and the CUDA backend then crashed on load with ptxas fatal: Memory allocation failure / CUDA error: a PTX JIT compilation failed.

If your desktop already uses a lot of VRAM, lower the context: /set parameter num_ctx 65536.

Chat template fixes

Unsloth's template raises errors that break popular agent tools. Two changes (full diff in chat_template.fixes.diff):

System messages after the start of the conversation. The template raised System message must be at the beginning. Claude Code (via Ollama's Anthropic-compatible /v1/messages) sends these, so every request failed with HTTP 500. Late system messages are now rendered as normal <|im_start|>system turns.
Unknown reasoning efforts. The template only accepted xhigh, high, medium and low, and raised Unexpected reasoning effort max for anything else. Codex sends max. Now max, ultra and other unknown values map to xhigh, and minimal/none map to low.

Ollama's Modelfile TEMPLATE only accepts Go templates, so the Jinja template had to be rewritten inside the GGUF. tools/gguf_set_template.py does that with the standard library only, and checks that every other key and all tensor data are unchanged:

python tools/gguf_set_template.py original.gguf fixed.gguf chat_template.jinja

Run from this repo instead of the Ollama registry

hf download Jfish123/Qwen3.8-27B-16GB-GGUF --local-dir qwen3.8-coder
cd qwen3.8-coder
ollama create qwen3.8-coder -f Modelfile

llama.cpp:

llama-server -m Qwen3.8-27B-UD-IQ3_XXS-fixed.gguf -c 125000 -ngl 99 -fa on -ctk q4_0 -ctv q4_0

Benchmark

benchmark/lcb_score.py runs all 175 problems of LiveCodeBench v6 against a local Ollama model, grades each one against its hidden tests, and prints a pass@1 score. It saves results as it goes, so a stopped run resumes where it left off.

Score for this quant: run in progress.

Published references, at full precision (from the Qwen3.8-27B model card): Qwen3.8-27B 90.3, Claude Opus 4.6 Max 88.8. This harness is close to, but not identical to, the official one, so treat differences of a few points as noise.

Credits

Qwen team for Qwen3.8-27B (Apache 2.0)
Unsloth for the dynamic IQ3_XXS quantization

Downloads last month: 266

GGUF

Model size

27B params

Architecture

qwen35

Hardware compatibility

3-bit

Model tree for Jfish123/Qwen3.8-27B-16GB-GGUF

Base model

Qwen/Qwen3.8-27B

Quantized

(1640)

this model