Instructions to use hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF",
	filename="Qwen3-Coder-30B-A3B-asym-2bitexp.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 hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF
# Run inference directly in the terminal:
llama-cli -hf hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF
# Run inference directly in the terminal:
llama-cli -hf hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-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 hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF
# Run inference directly in the terminal:
./llama-cli -hf hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-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 hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF
# Run inference directly in the terminal:
./build/bin/llama-cli -hf hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF

Use Docker

docker model run hf.co/hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF

LM Studio
Jan

vLLM

How to use hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-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": "hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF

Ollama
How to use hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF with Ollama:
```
ollama run hf.co/hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF
```

Unsloth Studio

How to use hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-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 hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-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 hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF to start chatting

How to use hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-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": "hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-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 hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF with Docker Model Runner:
```
docker model run hf.co/hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF
```

Lemonade

How to use hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF

Run and chat with the model

lemonade run user.Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF-{{QUANT_TAG}}

List all available models

lemonade list

Qwen3-Coder-30B-A3B — Asymmetric 2-bit-Expert GGUF (imatrix)

An asymmetric, expert-aware quantization of Qwen/Qwen3-Coder-30B-A3B-Instruct (arch qwen3moe, 128 routed experts, 8 active per token, ~3B active params).

The idea (the "antirez" insight): in a routed-MoE model the bulk of the weights live in the expert FFNs, and most experts are only sparsely active. Push the routed experts to 2-bit where the model is most redundant, keep the attention and the embedding/output tied weights at higher precision where error is most damaging, and steer the per-tensor bit-allocation with an importance matrix (imatrix). The result fits comfortably in 16 GB with only a modest perplexity cost versus the standard 4-bit baseline.

Asymmetric quantization scheme

Tensor group	Type	Bits	Count
Routed expert gate (`ffn_gate_exps`)	`IQ2_S`	~2.5	48
Routed expert up (`ffn_up_exps`)	`IQ2_S`	~2.5	48
Routed expert down (`ffn_down_exps`)	`IQ3_S`	~3.44	48
Attention `q/k/v/output`	`Q4_K`	~4.5	192
`token_embd`	`Q6_K`	~6.6	1
`output` (lm_head)	`Q6_K`	~6.6	1

Notes:

down experts get an extra bit (IQ3_S) — they are more error-sensitive than gate/up, so they are protected.
This model architecture has no shared expert (expert_shared_feed_forward_length = 0), so there is no always-on expert to hold at high precision — all FFN experts are routed.
Quantization was guided by an imatrix computed over bartowski/calibration_datav3.txt (128 chunks, ctx 512). imatrix.dat is included in this repo.

Effective rate: 2.99 BPW, on-disk 11.4 GB (10.64 GiB).

Quality (perplexity, wikitext-2 raw test, 200 chunks @ ctx 512)

Model	PPL	Δ vs Q4_K_M
This asym 2-bit-expert (2.99 BPW, 11.4 GB)	9.93	+0.33 (+3.4%)
Standard `Q4_K_M` (~4.8 BPW, 18.6 GB)	9.60	—

PPL measured with the same harness and chunk count for both. Lower is better. The asym build trades a small PPL increase for a ~39% smaller file that clears the 16 GB bar.

16 GB fit

Weights on disk / in VRAM: 11.4 GB.
KV cache (this arch: 48 layers, GQA, n_head_kv small) at f16 is on the order of ~0.13 GB per 1K tokens, so a 16K-token context adds roughly ~2 GB.
11.4 GB weights + 2 GB KV (16K ctx) + runtime overhead ≈ **14 GB < 16 GB**. ✅

Use a quantized KV cache (-ctk q8_0 -ctv q8_0) to push context further.

Usage (llama.cpp)

# This is the Instruct (non-thinking) variant — no <think> blocks.
llama-server -m Qwen3-Coder-30B-A3B-asym-2bitexp.gguf -ngl 99 -c 16384

Provenance / reproducibility

Source: unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF at Q8_0 (near-lossless) as the requantization source (--allow-requantize).
imatrix corpus: bartowski/calibration_datav3.txt, 128 chunks @ ctx 512.
Tooling: llama-quantize with repeatable --tensor-type REGEX=TYPE overrides plus --token-embedding-type Q6_K --output-tensor-type Q6_K, base type IQ3_S, imatrix-guided.

Coherence verified on a coding task (correct merge_intervals) and a multi-step word problem (35 heads / 94 legs → 23 chickens, 12 rabbits, with a correct check).

Downloads last month: 238

GGUF

Model size

31B params

Architecture

qwen3moe

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Model tree for hyperspaceai/Qwen3-Coder-30B-A3B-asym-2bitexp-GGUF

Base model

Qwen/Qwen3-Coder-30B-A3B-Instruct

Quantized

(148)

this model