Instructions to use MatterWaveAI/grt-30m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MatterWaveAI/grt-30m with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="MatterWaveAI/grt-30m")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("MatterWaveAI/grt-30m", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use MatterWaveAI/grt-30m 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 MatterWaveAI/grt-30m:F32
# Run inference directly in the terminal:
llama cli -hf MatterWaveAI/grt-30m:F32

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf MatterWaveAI/grt-30m:F32
# Run inference directly in the terminal:
llama cli -hf MatterWaveAI/grt-30m:F32

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 MatterWaveAI/grt-30m:F32
# Run inference directly in the terminal:
./build/bin/llama cli -hf MatterWaveAI/grt-30m:F32

Use Docker

docker model run hf.co/MatterWaveAI/grt-30m:F32

Llama new
LM Studio
Jan

vLLM

How to use MatterWaveAI/grt-30m with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MatterWaveAI/grt-30m"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MatterWaveAI/grt-30m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/MatterWaveAI/grt-30m:F32

SGLang

How to use MatterWaveAI/grt-30m with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "MatterWaveAI/grt-30m" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MatterWaveAI/grt-30m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "MatterWaveAI/grt-30m" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MatterWaveAI/grt-30m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use MatterWaveAI/grt-30m with Ollama:
```
ollama run hf.co/MatterWaveAI/grt-30m:F32
```
Unsloth Desktop
Docker Model Runner
How to use MatterWaveAI/grt-30m with Docker Model Runner:
```
docker model run hf.co/MatterWaveAI/grt-30m:F32
```

Lemonade

How to use MatterWaveAI/grt-30m with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull MatterWaveAI/grt-30m:F32

Run and chat with the model

lemonade run user.grt-30m-F32

List all available models

lemonade list

Atomic Chat

davidhouseholter commited on about 18 hours ago

Commit

1d63de9

verified ·

1 Parent(s): 54cb2d3

Update MatterWaveAI/grt-30m-curriculum final 24h metrics in README.md

Browse files

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -22,7 +22,7 @@ pipeline_tag: text-generation
 # GRT-30M (24-Hour Qwen 3 Distillation Preview)
-> **Multi-Round Curriculum Checkpoint Available:** For the multi-round `HuygensSymplecticAdam` curriculum adaptation of `GRT-30M` (**`47.16%`** 9-task `SLM-Bench` multiple-choice macro average, **`99.20%`** `GSM8K` 4-way verification accuracy, **`7.60`** `GSM8K` perplexity), see **[`MatterWaveAI/grt-30m-curriculum`](https://huggingface.co/MatterWaveAI/grt-30m-curriculum)**.
 A **Geometric Rotation Transformer (`GRTForCausalLM`)** distilled from **Qwen 3** (sharing the `151,669`-token Qwen 3 BPE vocabulary) in a single **24-hour distillation run** (`10.79B` tokens at `~123.9k tok/s` across `8x NVIDIA A100-SXM4-80GB`). GRT replaces every dense linear projection (`nn.Linear`) in attention (`Q, K, V, O`) and feed-forward (`expand, contract`) with cascades of content-modulated `2×2` Givens rotations on `SO(D)` interleaved with butterfly permutations.
@@ -132,7 +132,7 @@ Evaluated across the 10-task `SLM-Bench` core suite on an AMD Ryzen AI Max+ 395
 | Metric | `GRT-30M` (Zero-Shot) | `PairAdam` | `SymplecticAdam` | `HuygensSymplecticAdam` | Improvement (`HuygensSymplectic` vs. Zero-Shot) |
 |---|---:|---:|---:|---:|---:|
-| **GSM8K Held-Out Perplexity (PPL)** ↓ | `28.00` | `10.60` | `11.00` | **`9.04`** | **`-18.96` PPL (`3.10x` lower)** |
 | **GSM8K Solution Verification Accuracy (4-way MC)** ↑ | `34.20%` | `39.20%` | `98.40%` | **`98.60%`** | **`+64.40%`** |
 ### 2. Inference & Kernel Throughput on AMD Strix Halo (`Radeon 8060S` / `gfx1151`)

 # GRT-30M (24-Hour Qwen 3 Distillation Preview)
+> **Multi-Round Curriculum Checkpoint Available:** For the multi-round `HuygensSymplecticAdam` curriculum adaptation of `GRT-30M` (**`48.07%`** 9-task `SLM-Bench` multiple-choice macro average, **`99.20%`** `GSM8K` 4-way verification accuracy, **`7.60`** `GSM8K` perplexity), see **[`MatterWaveAI/grt-30m-curriculum`](https://huggingface.co/MatterWaveAI/grt-30m-curriculum)**.
 A **Geometric Rotation Transformer (`GRTForCausalLM`)** distilled from **Qwen 3** (sharing the `151,669`-token Qwen 3 BPE vocabulary) in a single **24-hour distillation run** (`10.79B` tokens at `~123.9k tok/s` across `8x NVIDIA A100-SXM4-80GB`). GRT replaces every dense linear projection (`nn.Linear`) in attention (`Q, K, V, O`) and feed-forward (`expand, contract`) with cascades of content-modulated `2×2` Givens rotations on `SO(D)` interleaved with butterfly permutations.
 | Metric | `GRT-30M` (Zero-Shot) | `PairAdam` | `SymplecticAdam` | `HuygensSymplecticAdam` | Improvement (`HuygensSymplectic` vs. Zero-Shot) |
 |---|---:|---:|---:|---:|---:|
+| **GSM8K Held-Out Perplexity (PPL)** ↓ | `27.99` | `10.65` | `10.98` | **`9.04`** | **`-18.95` PPL (`3.10x` lower)** |
 | **GSM8K Solution Verification Accuracy (4-way MC)** ↑ | `34.20%` | `39.20%` | `98.40%` | **`98.60%`** | **`+64.40%`** |
 ### 2. Inference & Kernel Throughput on AMD Strix Halo (`Radeon 8060S` / `gfx1151`)