Instructions to use madebywest/maple-pro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use madebywest/maple-pro 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 madebywest/maple-pro
# Run inference directly in the terminal:
llama cli -hf madebywest/maple-pro

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf madebywest/maple-pro
# Run inference directly in the terminal:
llama cli -hf madebywest/maple-pro

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 madebywest/maple-pro
# Run inference directly in the terminal:
./build/bin/llama cli -hf madebywest/maple-pro

Use Docker

docker model run hf.co/madebywest/maple-pro

Llama new
LM Studio
Jan
Ollama
How to use madebywest/maple-pro with Ollama:
```
ollama run hf.co/madebywest/maple-pro
```
Unsloth Desktop

How to use madebywest/maple-pro with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf madebywest/maple-pro

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": "madebywest/maple-pro"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use madebywest/maple-pro with Docker Model Runner:
```
docker model run hf.co/madebywest/maple-pro
```

Lemonade

How to use madebywest/maple-pro with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull madebywest/maple-pro

Run and chat with the model

lemonade run user.maple-pro-{{QUANT_TAG}}

List all available models

lemonade list

Hermes Agent

How to use madebywest/maple-pro with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf madebywest/maple-pro

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 madebywest/maple-pro

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use madebywest/maple-pro with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf madebywest/maple-pro

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 "madebywest/maple-pro" \
  --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"

maple-pro / README.md

madebywest

Upload 2 files

0b0f4bc verified 3 days ago

preview code

Raw History Blame Contribute Delete

5.06 kB

metadata

license: mit
base_model:
  - deepgrove/maple-preview
tags:
  - maple
  - gguf
  - dpo
  - preference-optimization
  - on-policy-distillation
  - moe
  - ternary
  - apple-silicon
library_name: gguf

Maple Pro — preference-optimized Maple (20B-A1B ternary MoE)

A single-file GGUF of Maple (deepgrove/maple-preview, MIT) after a full distillation + DPO campaign: the strongest measured variant of this model family, improving or matching the baseline on all 12 evaluated axes.

Results (measured on a fixed local suite, identical seeds/conditions)

Axis	Base Maple	Maple Pro
Coding probe (26 tasks)	20	24
GSM8K (pooled n=200)	90/100	90.5/100
HumanEval+ (50)	44	44
IFEval official checker (100)	63	65
Tools eval (12)	11	12/12
Knowledge (25)	22	23
Browser-agent probe (10)	3	5
IF-10 / IF-30	7 / 19	8 / 19
DROP-30	20	20
Median tokens/answer	759	570 (−25%)

First and only configuration (of 17 tested) achieving simultaneous dominance across every axis. All numbers reproducible; eval suite and logs described in the model card below.

What's inside

Full LoRA fusion: 64 attention LoRA tensors and 121 DPO-updated base tensors (norms, expert routers, QK-norms) baked into a single GGUF. Attention projections stored as F16 (mixed precision; layers 0–7 keep the base ternary TQ2_0 attention); ternary expert weights untouched (byte-identical to the official base).
Runtime: 6.6 GB, ~1B active parameters.
Context: 128K native, GGUF metadata allows 256K — verified to work natively with no flags (needle-tested — see Context section). No other changes to the model: the 256K build differs from the 128K original by a single metadata field.

Context (measured)

5-needle recall (wikitext-2 haystack, depths 10/30/50/70/90%), same seeds:

Context	Mode	Recall
16K–131K	native	3/5
256K	native (no flags)	4/5
256K	YaRN ×2	4/5 (same)
384K	YaRN ×3	2/5 — fails

No length degradation up to 256K: the hybrid SWA design (18/24 layers at window 512) extrapolates ×2 natively, so 256K works by just setting the context size — no rope-scaling flags needed (LM Studio: set the slider). Remaining misses are section-binding confusions between same-shape needles, uniform across lengths. 384K+ is out of reach without long-context training.

Runtime (important)

The maple architecture is not in upstream llama.cpp. Use the DeepGrove fork (Apache-2.0):

git clone --depth 1 https://github.com/deepgrove-ai/llama.cpp
cmake -B build -DCMAKE_BUILD_TYPE=Release
cmake --build build -j --target llama-server

The Metal backend of the current fork does not implement the ternary TQ2_0 matmuls (experts, and the ternary attention of layers 0–7), so serve with those tensors on CPU:

./build/bin/llama-server -m maple-pro.gguf --port 8080 -t 8 \
  --cpu-moe --no-op-offload \
  -ot 'blk\.[0-7]\.attn_[a-z]+\.weight=CPU'

OpenAI-compatible API on :8080. LM Studio works too: the GGUF was verified on LM Studio's bundled runtime (2.65.1) — just load it and set the context size. Measured decode on an M4 16 GB (idle): ~75 tok/s (base maple-preview: ~94 tok/s — Pro trades some bandwidth for its F16 fused attention).

Serving at 256K

The GGUF advertises a 262144 context: just set the context size to 256K — no rope-scaling flags required (native ×2 extrapolation is needle-verified, see Context section). On a 16 GB machine use quantized KV cache to fit:

./build/bin/llama-server -m maple-pro.gguf --port 8080 -t 8 \
  --ctx-size 262144 \
  --cpu-moe --no-op-offload \
  -ot 'blk\.[0-7]\.attn_[a-z]+\.weight=CPU' \
  --cache-type-k q8_0 --cache-type-v q8_0

KV cache quantization is your choice; q8_0 was used in the tests (~10 GB total on a 16 GB machine).

Training recipe (summary)

Semi on-policy distillation: the student samples its own solutions on verified tasks (GSM8K gold answers, IFEval-style constraint checkers, sandboxed coding tests); a frontier teacher API model (GLM-5.3-Flash) corrects only the failures. ~3,000 verified on-policy samples + corrections.
SFT on-policy multi-domain (the 11/12-axes intermediate).
DPO on preference pairs (student error ↔ verified teacher correction), full-parameter (attention LoRA + norms/routers), SFT-anchored.
Zero-tolerance gate: a candidate ships only if it matches or beats the baseline on all axes at once. 16 configurations were rejected by this gate before this one passed.

Credits & license

Base model: deepgrove/maple-preview and its GGUF release (MIT).
Runtime: deepgrove-ai/llama.cpp fork.
This model: MIT, derivative work with attribution. Trained locally on a single M4 16 GB via MLX.
Made by West — madebywest.com