Instructions to use PollardWeights/Qwen3.8-27B-Pollard with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use PollardWeights/Qwen3.8-27B-Pollard 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 PollardWeights/Qwen3.8-27B-Pollard:IQ3_S
# Run inference directly in the terminal:
llama cli -hf PollardWeights/Qwen3.8-27B-Pollard:IQ3_S

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf PollardWeights/Qwen3.8-27B-Pollard:IQ3_S
# Run inference directly in the terminal:
llama cli -hf PollardWeights/Qwen3.8-27B-Pollard:IQ3_S

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 PollardWeights/Qwen3.8-27B-Pollard:IQ3_S
# Run inference directly in the terminal:
./llama-cli -hf PollardWeights/Qwen3.8-27B-Pollard:IQ3_S

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 PollardWeights/Qwen3.8-27B-Pollard:IQ3_S
# Run inference directly in the terminal:
./build/bin/llama-cli -hf PollardWeights/Qwen3.8-27B-Pollard:IQ3_S

Use Docker

docker model run hf.co/PollardWeights/Qwen3.8-27B-Pollard:IQ3_S

LM Studio
Jan

vLLM

How to use PollardWeights/Qwen3.8-27B-Pollard with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "PollardWeights/Qwen3.8-27B-Pollard"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "PollardWeights/Qwen3.8-27B-Pollard",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/PollardWeights/Qwen3.8-27B-Pollard:IQ3_S

Ollama
How to use PollardWeights/Qwen3.8-27B-Pollard with Ollama:
```
ollama run hf.co/PollardWeights/Qwen3.8-27B-Pollard:IQ3_S
```

Unsloth Studio

How to use PollardWeights/Qwen3.8-27B-Pollard 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 PollardWeights/Qwen3.8-27B-Pollard 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 PollardWeights/Qwen3.8-27B-Pollard to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for PollardWeights/Qwen3.8-27B-Pollard to start chatting

How to use PollardWeights/Qwen3.8-27B-Pollard with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf PollardWeights/Qwen3.8-27B-Pollard:IQ3_S

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": "PollardWeights/Qwen3.8-27B-Pollard:IQ3_S"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use PollardWeights/Qwen3.8-27B-Pollard with Docker Model Runner:
```
docker model run hf.co/PollardWeights/Qwen3.8-27B-Pollard:IQ3_S
```

Lemonade

How to use PollardWeights/Qwen3.8-27B-Pollard with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull PollardWeights/Qwen3.8-27B-Pollard:IQ3_S

Run and chat with the model

lemonade run user.Qwen3.8-27B-Pollard-IQ3_S

List all available models

lemonade list

Hermes Agent

How to use PollardWeights/Qwen3.8-27B-Pollard with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf PollardWeights/Qwen3.8-27B-Pollard:IQ3_S

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 PollardWeights/Qwen3.8-27B-Pollard:IQ3_S

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use PollardWeights/Qwen3.8-27B-Pollard with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf PollardWeights/Qwen3.8-27B-Pollard:IQ3_S

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 "PollardWeights/Qwen3.8-27B-Pollard:IQ3_S" \
  --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"

Pollard quantizations of Qwen3.8-27B by Qwen

The Qwen3.8-27B flagship — 27B, multimodal (text + image + video), 256K context — sized to your box. The recommended build is ~13 GB and runs on a 16 GB Mac; a 24 GB / 32 GB tier is there when you have the room. Built with Pollard Weights on a runtime with native qwen35 support, using Qwen3.8-27B's own imatrix (exact match).

Original model: https://huggingface.co/Qwen/Qwen3.8-27B

Model details


Parameter count	27B (dense)
Architecture	`qwen35` (65 layers, 5120 hidden)
Context length	262,144 (256K)
Input support	text + image + video (multimodal)
imatrix	yes — Qwen3.8-27B's own importance matrix (exact)
Source	Q8_0 (near-lossless; ~99.9% of bf16)

Which file should I choose?

~16 GB Mac / 12 GB VRAM → IQ3_S (~13 GB). Recommended.
~24 GB (a 3090) → IQ4_XS (~16 GB) — more fidelity.
~32 GB → Q6_K (~20 GB) — near-lossless.

Available files

file	quant	size	held-out PPL	M4 tok/s
Qwen3.8-27B-Pollard-IQ3_S.gguf	IQ3_S mix	~13 GB	(see repo)	(16 GB tier)
Qwen3.8-27B-Pollard-IQ4_XS.gguf	IQ4_XS mix	~16 GB	—	24 GB tier
Qwen3.8-27B-Pollard-Q6_K.gguf	Q6_K mix	~20 GB	—	32 GB tier
mmproj-Qwen3.8-27B-bf16.gguf	vision projector (bf16, not quantized)	0.93 GB	—	ship w/ any

(PPL/tok-s measured only on the IQ3 tier — the larger tiers exceed a 16 GB Mac's Metal budget, which is the point: those are for 24 GB / 32 GB boxes.)

Prompt format

ChatML (Qwen3.5):

<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

Multimodal (vision + video)

Ship the mmproj alongside the model (kept at bf16 — never quantize the projector):

llama-mtmd-cli -m Qwen3.8-27B-Pollard-IQ3_S.gguf \
  --mmproj mmproj-Qwen3.8-27B-bf16.gguf --image photo.jpg -p "Describe this image."
llama-server -m Qwen3.8-27B-Pollard-IQ3_S.gguf --mmproj mmproj-Qwen3.8-27B-bf16.gguf -ngl 99

Download & run

pip install -U "huggingface_hub[cli]"
hf download PollardWeights/Qwen3.8-27B-Pollard \
  --include "Qwen3.8-27B-Pollard-IQ3_S.gguf" "mmproj-Qwen3.8-27B-bf16.gguf" --local-dir ./
llama-server -m Qwen3.8-27B-Pollard-IQ3_S.gguf -ngl 99   # OpenAI API + web UI :8080

Needs a recent llama.cpp with qwen35 support.

Notes

License: Apache-2.0, inherited from Qwen3.8-27B.
Requantized from Q8_0 (near-lossless source; negligible vs bf16 for an IQ3/IQ4 target).
imatrix is Qwen3.8-27B's own — imatrix-uncovered tensors pinned to q6_K.
Quantized, not fine-tuned — same weights, sized to your machine.

Credits

Base model: Qwen/Qwen3.8-27B (Qwen team).
Quantization runtime: llama.cpp.
Method + tooling: Pollard Weights — measure first, no claim before a number.

Downloads last month: 14

GGUF

Model size

27B params

Architecture

qwen35

Hardware compatibility

3-bit

4-bit

6-bit

Model tree for PollardWeights/Qwen3.8-27B-Pollard

Base model

Qwen/Qwen3.8-27B

Quantized

(830)

this model