Deepseek R1
DeepSeek, the creator, develops the DeepSeek R 1 model, a chat-type AI capable of code generation, math, reasoning, streaming, and text generation. It is genuinely best at tasks involving math and reasoning, as evidenced by its top 25% benchmark scores in MMLU-Pro and MATH-500. Notably, DeepSeek R 1 has a large context window of 65,536 tokens, allowing it to process and understand extensive inputs.
Input
Output
Context
164K
Max Output
8K
Parameters
684.5B
Technical Specifications
Model TypeChat
Context Window163,840 tokens
Max Output Tokens8,192 tokens
Parameters684.5B
Release DateJan 20, 2025
Training CutoffOct 1, 2023
Licensemit
Open SourceYes
Input Modalities
Text
Output Modalities
Text
Capabilities
Benchmarks
Artificial Analysis
9.3%
HLE
68.3%
AIME
70.8%
GPQA
35.7%
SciCode
96.6%
MATH 500
84.4%
MMLU Pro
68.0
Math
15.9
Coding
61.7%
LiveCodeBench
12.6
Intelligence
Resources & Links
Related Tools
Data sourced from official provider APIs and documentation
Last updated: Jun 24, 2026
Start building with the right model.
Automatically route workloads to the right model for every task, every time.