DeepSeek-R1 70B
Parameters
70B
Context Length
33K
Modality
Text
Architecture
Dense
License
MIT License
Release Date
27 Dec 2024
Knowledge Cutoff
-
API Pricing (per 1M)
Input: $0.70 · Output: $1.10
System Requirements
VRAM requirements for different quantization methods and context sizes
1,024 tokens
Consumer
8x RTX 4090
24GB VRAM
Datacenter
3x NVIDIA A100
80GB VRAM
Apple Silicon
2x Apple M3 Max
128GB VRAM
32,768 tokens
Consumer
16x RTX 4090
24GB VRAM
Datacenter
5x NVIDIA A100
80GB VRAM
Apple Silicon
3x Apple M3 Max
128GB VRAM
Architecture Diagram
Evaluation Benchmarks
Rank
#139
| Benchmark | Score | Rank |
|---|---|---|
Graduate-Level QA | 0.652 | 90 |
Intelligence Index | 0.04 | 189 |
Rankings
Overall Rank
#139
Coding Rank
-
About DeepSeek-R1 70B
DeepSeek-R1 70B is a dense reasoning foundation model distilled from DeepSeek-R1 onto the Llama-3.3-70B-Instruct architecture. It combines deep logical inference and competitive programming skills with efficient enterprise deployment profiles.
Technical Specifications
Attention
Attention Structure
Multi-Layer Attention
Attention Heads
112
Key-Value Heads
112
Attention Head Dimension
128
Position Embedding
ROPE
RoPE Theta
500,000
Sliding Window Attention
No
Sliding Window Size
-
Sliding Window Ratio
-
Linear Attention
-
Linear Attention Ratio
-
Normalization
RMS Normalization
Activation Function
Swish
Dimensions
Hidden Dimension Size
8,192
Number of Layers
80
FFN Intermediate Size (Dense)
28,672
Multi-Token Prediction Heads
-
Tokenizer
Vocabulary Size
128,256
About DeepSeek-R1
DeepSeek-R1 is a model family developed for logical reasoning tasks. It incorporates a Mixture-of-Experts architecture for computational efficiency and scalability. The family utilizes Multi-Head Latent Attention and employs reinforcement learning in its training, with some variants integrating cold-start data.