Instructions to use OpenMOSE/HRWKV7-Reka-Flash3-Preview with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OpenMOSE/HRWKV7-Reka-Flash3-Preview with Transformers:

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

pipe = pipeline("text-generation", model="OpenMOSE/HRWKV7-Reka-Flash3-Preview")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("OpenMOSE/HRWKV7-Reka-Flash3-Preview", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use OpenMOSE/HRWKV7-Reka-Flash3-Preview with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OpenMOSE/HRWKV7-Reka-Flash3-Preview"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenMOSE/HRWKV7-Reka-Flash3-Preview",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/OpenMOSE/HRWKV7-Reka-Flash3-Preview

SGLang

How to use OpenMOSE/HRWKV7-Reka-Flash3-Preview 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 "OpenMOSE/HRWKV7-Reka-Flash3-Preview" \
    --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": "OpenMOSE/HRWKV7-Reka-Flash3-Preview",
		"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 "OpenMOSE/HRWKV7-Reka-Flash3-Preview" \
        --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": "OpenMOSE/HRWKV7-Reka-Flash3-Preview",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use OpenMOSE/HRWKV7-Reka-Flash3-Preview with Docker Model Runner:
```
docker model run hf.co/OpenMOSE/HRWKV7-Reka-Flash3-Preview
```

OpenMOSE commited on Jul 6, 2025

Commit

88fb0d7

verified ·

1 Parent(s): 3bb3693

Update README.md

Browse files

Files changed (1) hide show

README.md +4 -0

README.md CHANGED Viewed

@@ -7,6 +7,10 @@ license: apache-2.0
   <img src="./hxa079.png" style="border-radius: 15px; width: 60%; height: 60%; object-fit: cover;  box-shadow: 10px 10px 20px rgba(0, 0, 0, 0.5); border: 2px solid white;" alt="PRWKV" />
 </div>
 ### Model Description
 HRWKV7-Reka-Flash3-Preview is an experimental hybrid architecture model that combines RWKV v7's linear attention mechanism with Group Query Attention (GQA) layers. Built upon the Reka-flash3 21B foundation, this model replaces most Transformer attention blocks with RWKV blocks while strategically maintaining some GQA layers to enhance performance on specific tasks.

   <img src="./hxa079.png" style="border-radius: 15px; width: 60%; height: 60%; object-fit: cover;  box-shadow: 10px 10px 20px rgba(0, 0, 0, 0.5); border: 2px solid white;" alt="PRWKV" />
 </div>
+> I'm simply exploring the possibility of linearizing existing Transformer models.
+> It's still far from perfect,
+> but I hope you'll bear with me as I continue this journey.
 ### Model Description
 HRWKV7-Reka-Flash3-Preview is an experimental hybrid architecture model that combines RWKV v7's linear attention mechanism with Group Query Attention (GQA) layers. Built upon the Reka-flash3 21B foundation, this model replaces most Transformer attention blocks with RWKV blocks while strategically maintaining some GQA layers to enhance performance on specific tasks.