Instructions to use Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT",
	filename="mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00001-of-00019.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R
# Run inference directly in the terminal:
llama-cli -hf Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R
# Run inference directly in the terminal:
llama-cli -hf Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R

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 Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R
# Run inference directly in the terminal:
./llama-cli -hf Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R

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 Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R

Use Docker

docker model run hf.co/Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R

LM Studio
Jan
Ollama
How to use Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT with Ollama:
```
ollama run hf.co/Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R
```

Unsloth Studio

How to use Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT 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 Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT 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 Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT to start chatting

How to use Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R

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": "Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R

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 Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT with Docker Model Runner:
```
docker model run hf.co/Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R
```

Lemonade

How to use Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Thireus/mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT:IQ2_XXS_R

Run and chat with the model

lemonade run user.mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT-IQ2_XXS_R

List all available models

lemonade list

mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT / tensors.map

Thireus

mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_SPLIT

a674e6e 23 days ago

raw

history blame contribute delete

3.96 kB

	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00002-of-00019.gguf:b9f047b03ba808b2d8d5296475240f781b66b3b6a264b93698ffbf397171030a:output.weight:shape=(5120, 248320):dtype=iq2_xxs_r4:elements=1271398400:bytes=327782400
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00003-of-00019.gguf:482e9c070491ce9fb49aba03e2f2f6496dc45cac2cdb4a93230192dfa3a372f7:token_embd.weight:shape=(5120, 248320):dtype=iq2_xxs:elements=1271398400:bytes=327782400
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00004-of-00019.gguf:6f87290abdd9785ba8d6787ba04394721aba5968c9cb82b79234a13eff56726d:blk.64.ffn_down.weight:shape=(17408, 5120):dtype=iq2_xxs_r4:elements=89128960:bytes=22978560
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00005-of-00019.gguf:4257ffee3db9baa1dffff752f13505255779c787e7dbbbfa39623dd871afefc3:blk.64.ffn_gate.weight:shape=(5120, 17408):dtype=iq2_xxs_r4:elements=89128960:bytes=22978560
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00006-of-00019.gguf:971cc66ac7bed0d96fdead32d51a8cdf0d5880f960dae7cadc1a51478b9bc106:blk.64.ffn_up.weight:shape=(5120, 17408):dtype=iq2_xxs_r4:elements=89128960:bytes=22978560
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00007-of-00019.gguf:bc47a2ef78a53d8169cd75d3b8f77441ff11ba4529c9e3bd4f7b57bde4d4a076:blk.64.nextn.eh_proj.weight:shape=(10240, 5120):dtype=iq2_xxs_r4:elements=52428800:bytes=13516800
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00008-of-00019.gguf:cffe2291e3aff9b242e5676ef065c9b7d842b97b0acd617ffe0226ffd8c4991a:blk.64.attn_q.weight:shape=(5120, 12288):dtype=iq2_xxs_r4:elements=62914560:bytes=16220160
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00009-of-00019.gguf:b46321af218a4ee256322cf980d692e4724abad108a1dc4722394fda0a25dbd7:blk.64.attn_output.weight:shape=(6144, 5120):dtype=iq2_xxs_r4:elements=31457280:bytes=8110080
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00010-of-00019.gguf:4590cfcb507003c61300b1abfd64d9e98e2fa3091d115f4fc5876a11c6851023:output_norm.weight:shape=(5120,):dtype=f32:elements=5120:bytes=20480
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00011-of-00019.gguf:57424507743ab0f420971c731168c70d6a78436f4d9162abd15bb0ea8f5f6617:blk.64.attn_norm.weight:shape=(5120,):dtype=f32:elements=5120:bytes=20480
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00012-of-00019.gguf:d016a77a9d9258dcd2a2be3b3d2bc9625bb767e59d79914e5306033e3fbe40a8:blk.64.post_attention_norm.weight:shape=(5120,):dtype=f32:elements=5120:bytes=20480
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00013-of-00019.gguf:ce65b7f5b2027ce61069856bd806a44a66669dbf9e32f9d2b8786bba1d5ddb4e:blk.64.attn_k_norm.weight:shape=(256,):dtype=f32:elements=256:bytes=1024
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00014-of-00019.gguf:8800a55ffec2c2673b1b4b11d4f0aaf274e7474425ddb0ec598698416ad4916b:blk.64.attn_k.weight:shape=(5120, 1024):dtype=iq2_xxs_r4:elements=5242880:bytes=1351680
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00015-of-00019.gguf:3bb5caaa495668e3cd629734f66d0b42870a234ba6728467fc0a21c63a04741d:blk.64.attn_q_norm.weight:shape=(256,):dtype=f32:elements=256:bytes=1024
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00016-of-00019.gguf:f8974e532935a41f762007f9ffaddc9396c91b5ebed2e7ae6544c359c4a1b0eb:blk.64.attn_v.weight:shape=(5120, 1024):dtype=iq2_xxs_r4:elements=5242880:bytes=1351680
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00017-of-00019.gguf:f2b8b5abd6e5b248120fae9f23ef73bb6c9959ff6914bcf320ea78f4c9b71ebc:blk.64.nextn.shared_head_norm.weight:shape=(5120,):dtype=f32:elements=5120:bytes=20480
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00018-of-00019.gguf:b8fd0bde64470d9653e9846cf0515aef9771d83261961a6e08fdb4bc5b08c385:blk.64.nextn.enorm.weight:shape=(5120,):dtype=f32:elements=5120:bytes=20480
	mtp-Qwen3.5-27B-THIREUS-IQ2_XXS_R4-SPECIAL_TENSOR-00019-of-00019.gguf:7edce47fd6bcefa830c40964540b2cde273e99dacdfcdb494da75beb01e98a6c:blk.64.nextn.hnorm.weight:shape=(5120,):dtype=f32:elements=5120:bytes=20480