Open-R1: a fully open reproduction of DeepSeek-R1
•
221
Lewis Tunstall PRO
lewtun
AI & ML interests
LLMs, LLMs, LLMs
Recent Activity
liked
a dataset
about 3 hours ago
ServiceNow-AI/R1-Distill-SFT
liked
a dataset
about 3 hours ago
open-thoughts/OpenThoughts-114k
upvoted
an
article
about 4 hours ago
Welcome to Inference Providers on the Hub 🔥
Articles
Organizations
lewtun's activity
liked
2
datasets
about 3 hours ago
upvoted
an
article
about 4 hours ago
Article
Welcome to Inference Providers on the Hub 🔥
•
103
commented on
Open-R1: a fully open reproduction of DeepSeek-R1
about 12 hours ago
Hi @bojan2501 thanks, we will indeed be working hard to make sure this training recipe can work for small language models on consumer hardware since not everyone has a cluster of H100s at home :)
The tool we used for the images was Excalidraw! https://excalidraw.com
upvoted
an
article
1 day ago
Article
Open-R1: a fully open reproduction of DeepSeek-R1
•
221
updated
a
dataset
1 day ago
reacted to
fdaudens's
post with 🔥❤️
1 day ago
Post
4175
Yes, DeepSeek R1's release is impressive. But the real story is what happened in just 7 days after:
- Original release: 8 models, 540K downloads. Just the beginning...
- The community turned those open-weight models into +550 NEW models on Hugging Face. Total downloads? 2.5M—nearly 5X the originals.
The reason? DeepSeek models are open-weight, letting anyone build on top of them. Interesting to note that the community focused on quantized versions for better efficiency & accessibility. They want models that use less memory, run faster, and are more energy-efficient.
When you empower builders, innovation explodes. For everyone. 🚀
The most popular community model? @bartowski 's DeepSeek-R1-Distill-Qwen-32B-GGUF version — 1M downloads alone.
- Original release: 8 models, 540K downloads. Just the beginning...
- The community turned those open-weight models into +550 NEW models on Hugging Face. Total downloads? 2.5M—nearly 5X the originals.
The reason? DeepSeek models are open-weight, letting anyone build on top of them. Interesting to note that the community focused on quantized versions for better efficiency & accessibility. They want models that use less memory, run faster, and are more energy-efficient.
When you empower builders, innovation explodes. For everyone. 🚀
The most popular community model? @bartowski 's DeepSeek-R1-Distill-Qwen-32B-GGUF version — 1M downloads alone.
- 2 replies
published
a
dataset
2 days ago
updated
a
dataset
2 days ago
posted
an
update
3 days ago
Post
8102
We are reproducing the full DeepSeek R1 data and training pipeline so everybody can use their recipe. Instead of doing it in secret we can do it together in the open!
🧪 Step 1: replicate the R1-Distill models by distilling a high-quality reasoning corpus from DeepSeek-R1.
🧠 Step 2: replicate the pure RL pipeline that DeepSeek used to create R1-Zero. This will involve curating new, large-scale datasets for math, reasoning, and code.
🔥 Step 3: show we can go from base model -> SFT -> RL via multi-stage training.
Follow along: https://github.com/huggingface/open-r1
🧪 Step 1: replicate the R1-Distill models by distilling a high-quality reasoning corpus from DeepSeek-R1.
🧠 Step 2: replicate the pure RL pipeline that DeepSeek used to create R1-Zero. This will involve curating new, large-scale datasets for math, reasoning, and code.
🔥 Step 3: show we can go from base model -> SFT -> RL via multi-stage training.
Follow along: https://github.com/huggingface/open-r1
- 5 replies