Instructions to use recursal/radlads-7b-various with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use recursal/radlads-7b-various with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="recursal/radlads-7b-various")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("recursal/radlads-7b-various", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use recursal/radlads-7b-various with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "recursal/radlads-7b-various" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "recursal/radlads-7b-various", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/recursal/radlads-7b-various
- SGLang
How to use recursal/radlads-7b-various 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 "recursal/radlads-7b-various" \ --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": "recursal/radlads-7b-various", "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 "recursal/radlads-7b-various" \ --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": "recursal/radlads-7b-various", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use recursal/radlads-7b-various with Docker Model Runner:
docker model run hf.co/recursal/radlads-7b-various
Add Project Page Link and Github Link to Metadata (#4)
Browse files- Add Project Page Link and Github Link to Metadata (70197c3379eae934ef63f1835fb9478ce0826c85)
Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -1,7 +1,9 @@
|
|
| 1 |
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
library_name: transformers
|
|
|
|
| 4 |
pipeline_tag: text-generation
|
|
|
|
|
|
|
| 5 |
---
|
| 6 |
|
| 7 |
This repository contains various checkpoints for ablations and other unusual models from the paper [RADLADS: Rapid Attention Distillation to Linear Attention Decoders at Scale](https://huggingface.co/papers/2505.03005).
|
|
@@ -31,4 +33,4 @@ The file numbering is currently off by one from the step numbers shown in the pa
|
|
| 31 |
|L28-D3584-qwerky7_qwen2-3-4k-ckpt5.pth|2|Qwen2.5-7B-Instruct|RAD-RWKV7|4k ctxlen training, early checkpoint|
|
| 32 |
|L28-D3584-qwerky7_qwen2-3-4k.pth|2|Qwen2.5-7B-Instruct|RAD-RWKV7|4k ctxlen training|
|
| 33 |
|
| 34 |
-
More information can be found at the Github repository: https://github.com/recursal/RADLADS-paper
|
|
|
|
| 1 |
---
|
|
|
|
| 2 |
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
pipeline_tag: text-generation
|
| 5 |
+
project_page: https://sites.google.com/view/eagle-llm
|
| 6 |
+
repo_url: https://github.com/recursal/RADLADS-paper
|
| 7 |
---
|
| 8 |
|
| 9 |
This repository contains various checkpoints for ablations and other unusual models from the paper [RADLADS: Rapid Attention Distillation to Linear Attention Decoders at Scale](https://huggingface.co/papers/2505.03005).
|
|
|
|
| 33 |
|L28-D3584-qwerky7_qwen2-3-4k-ckpt5.pth|2|Qwen2.5-7B-Instruct|RAD-RWKV7|4k ctxlen training, early checkpoint|
|
| 34 |
|L28-D3584-qwerky7_qwen2-3-4k.pth|2|Qwen2.5-7B-Instruct|RAD-RWKV7|4k ctxlen training|
|
| 35 |
|
| 36 |
+
More information can be found at the Github repository: [https://github.com/recursal/RADLADS-paper](https://github.com/recursal/RADLADS-paper)
|