MDL
DeepSeek-R1-Distill-Llama-8B
deepseek-ai
Parameters
8B
Downloads
480.6K
on HF
Likes
862
Avg score
—
Weights
Open
Released
Jan 20, 2025
About
DeepSeek-R1-Distill-Llama-8B is a 8B-parameter open-weights model in the DeepSeek family from deepseek-ai.
mittext-generation
Benchmark scores · 0
No benchmark scores yet.
The research behind this model
all papers mentioning it →DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement LearningPhi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs4 cites · Mar 3, 2025Beyond a Single Direction: Chain-of-Thought Disrupts Simple Steering of RefusalMay 26, 2026When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for ReasoningMay 20, 2026The Coupling Tax: How Shared Token Budgets Undermine Visible Chain-of-Thought Under Fixed Output LimitsMay 8, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.