Instructions to use MainStack/marvy-2-35B-MoE-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MainStack/marvy-2-35B-MoE-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="MainStack/marvy-2-35B-MoE-GGUF",
	filename="marvy-2-35B-MoE-Q4_K_M.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 MainStack/marvy-2-35B-MoE-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 MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf MainStack/marvy-2-35B-MoE-GGUF: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 MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf MainStack/marvy-2-35B-MoE-GGUF: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 MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M

Use Docker

docker model run hf.co/MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use MainStack/marvy-2-35B-MoE-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MainStack/marvy-2-35B-MoE-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": "MainStack/marvy-2-35B-MoE-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M

Ollama
How to use MainStack/marvy-2-35B-MoE-GGUF with Ollama:
```
ollama run hf.co/MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M
```

Unsloth Studio

How to use MainStack/marvy-2-35B-MoE-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 MainStack/marvy-2-35B-MoE-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 MainStack/marvy-2-35B-MoE-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for MainStack/marvy-2-35B-MoE-GGUF to start chatting

How to use MainStack/marvy-2-35B-MoE-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf MainStack/marvy-2-35B-MoE-GGUF: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": "MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use MainStack/marvy-2-35B-MoE-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 MainStack/marvy-2-35B-MoE-GGUF: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 MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use MainStack/marvy-2-35B-MoE-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M

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 "MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M" \
  --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"

Docker Model Runner
How to use MainStack/marvy-2-35B-MoE-GGUF with Docker Model Runner:
```
docker model run hf.co/MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M
```

Lemonade

How to use MainStack/marvy-2-35B-MoE-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.marvy-2-35B-MoE-GGUF-Q4_K_M

List all available models

lemonade list

marvy-2-35B-MoE-GGUF

GGUF quants of marvy-2, a 35B-A3B Mixture-of-Experts model fine-tuned for the ServiceNow delivery lifecycle. ~3B active parameters per token; runs on a single consumer GPU or fast Apple Silicon thanks to the MoE sparsity.

GGUF quantizations for use with llama.cpp, Ollama, LM Studio, and compatible runtimes.

Released under Apache-2.0. Built with Qwen3.5 (Apache-2.0) via unsloth/Qwen3.6-35B-A3B and the Opus-distilled stamsam/...MTP base.

Files

File	Quant	Size	Use when
`marvy-2-35B-MoE-Q4_K_M.gguf`	Q4_K_M	~20 GB	Default — best size/quality balance
`marvy-2-35B-MoE-Q8_0.gguf`	Q8_0	~34 GB	Near-FP16 quality, more headroom

Architecture notes

This is a hybrid SSM + MoE Transformer:

40 layers, mixed SSM (Mamba-style) and grouped-query attention blocks
256 routed experts per MoE layer, 3B active per token (A3B)
Shared expert per layer for common pathways
qwen3_5_moe architecture in llama.cpp; needs a recent build (see "Supported runtimes" below)

The Multi-Token Prediction (MTP) head present in the base model was not included in this GGUF — these quants are text-only causal LM. The base model's MoE expert weights and SSM blocks are preserved.

Quick start

Ollama

ollama run hf.co/MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M

llama.cpp

llama-cli -hf MainStack/marvy-2-35B-MoE-GGUF:Q4_K_M \
  -p "Write a ServiceNow user story with acceptance criteria for P1 SLA escalation." \
  --temp 0.4 \
  -c 4096

LM Studio

Search the model catalog for marvy-2-35B-MoE-GGUF and download, or load the local .gguf file via "Open Model in Folder". LM Studio's OpenAI-compatible server is at http://localhost:1234/v1 by default.

Supported runtimes

The qwen3_5_moe architecture (with mixed SSM layers and Multi-Token Prediction in the base) is new. Verify your runtime supports it:

llama.cpp: master branch as of mid-2026 (commits containing Qwen3_5MoeForConditionalGeneration registration in conversion/qwen.py). The Homebrew formula may lag — clone upstream if you hit "unknown architecture" errors.
Ollama: ships its own llama.cpp; check that your Ollama version is recent.
LM Studio: uses its own bundled runtime; recent versions support Qwen3.5 MoE.

Trained on

v1 corpus: ServiceNow delivery lifecycle artifacts (SOW, SDD, stories, acceptance criteria, value hypothesis, ...) — same data marvy-1 used.
v2 corpus: extended to capability-to-epic mapping, mermaid diagram authoring, deployment package modeling, stakeholder mapping, story-to-UAT, and more. See EVAL.md in the repo root for per-task perplexity.

How this GGUF was built

LoRA adapter (rank 32, 350 steps, attention-only Q/K/V/O)
    +
bf16 base   (unsloth/Qwen3.6-35B-A3B)
    │   mlx_lm fuse
merged-bf16/  (14 shards, 65 GB safetensors with mlx_lm switch_mlp naming)
    │   scripts/marvy-v2-rename-moe-tensors.py  (bridge to HF-canonical names)
merged-bf16-hf/  (15 shards, 65 GB; switch_mlp → experts.gate_up_proj packed)
    │   llama.cpp/convert_hf_to_gguf.py --no-mtp --outtype f16
marvy-2-35B-MoE-F16.gguf  (65 GB, 733 tensors)
    │   llama-quantize
marvy-2-35B-MoE-Q4_K_M.gguf  (20 GB, 4.6 BPW)
marvy-2-35B-MoE-Q8_0.gguf    (34 GB, 8.52 BPW)

End-to-end build script: scripts/marvy-v2-35B-MoE-build-gguf.sh (in the source repo). The switch_mlp → experts rename is necessary because mlx_lm fuse and llama.cpp's converter use different naming conventions for routed MoE tensors.

License

Apache-2.0 (inherits from Qwen). See LICENSE and NOTICE in the repo root.

Citation

@misc{marvy-2-35B-MoE,
  title  = {marvy-2: A ServiceNow delivery-lifecycle MoE LLM},
  author = {MainStack},
  year   = {2026},
  howpublished = {\url{https://huggingface.co/MainStack/marvy-2-35B-MoE-GGUF}}
}

Downloads last month: 111

GGUF

Model size

35B params

Architecture

qwen35moe

Hardware compatibility

4-bit

8-bit

Model tree for MainStack/marvy-2-35B-MoE-GGUF

Base model

Qwen/Qwen3.6-35B-A3B

Quantized

stamsam/Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-MLX-oQ4-MTP

Quantized

(1)

this model