Scribd
Upload
6 views1 page

Into The Unknown: Start Now

DeepSeek-R1 is now live and open source, offering capabilities that compete with OpenAI's Model o1 across web, app, and API platforms. The new DeepSeek-V3 model boasts significant improvements in inference speed and ranks highly among both open-source and closed-source models. Users can access DeepSeek for free and utilize its various AI tools, including a mobile app for on-the-go chatting.

Uploaded by

georgephang12
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
6 views1 page

Into The Unknown: Start Now

DeepSeek-R1 is now live and open source, offering capabilities that compete with OpenAI's Model o1 across web, app, and API platforms. The new DeepSeek-V3 model boasts significant improvements in inference speed and ranks highly among both open-source and closed-source models. Users can access DeepSeek for free and utilize its various AI tools, including a mobile app for on-the-go chatting.

Uploaded by

georgephang12
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

API Platform ↗ 中⽂

DeepSeek-R1 is now live and open source,


rivaling OpenAI's Model o1. Available on web, app,
and API. Click for details.

Into the unknown

Start Now
Free access to DeepSeek-V3.
Experience the intelligent model.

Get DeepSeek App


Chat on the go with DeepSeek-V3
Your free all-in-one AI tool

https://www.deepseek.com/

DeepSeek-V3
Capabilities
DeepSeek-V3 achieves a significant
breakthrough in inference speed over
previous models.
It tops the leaderboard among open-source
models and rivals the most advanced
closed-source models globally.

DeepSeek
Benchmark (Metric)

Architecture MoE

# Activated Params 37B

# Total Params 671B

MMLU (EM) 88.5

MMLU-Redux (EM) 89.1

MMLU-Pro (EM) 75.9

DROP (3-shot F1) 91.6

English IF-Eval (Prompt Strict) 86.1

GPQA-Diamond (Pass@1) 59.1

SimpleQA (Correct) 24.9

FRAMES (Acc.) 73.3

LongBench v2 (Acc.) 48.7

HumanEval-Mul (Pass@1) 82.6

LiveCodeBench (Pass@1-COT) 40.5

LiveCodeBench (Pass@1) 37.6

Code Codeforces (Percentile) 51.6

SWE Verified (Resolved) 42.0

Aider-Edit (Acc.) 79.7

Aider-Polyglot (Acc.) 49.6

AIME 2024 (Pass@1) 39.2

Math MATH-500 (EM) 90.2

CNMO 2024 (Pass@1) 43.2

CLUEWSC (EM) 90.9

Chinese C-Eval (EM) 86.5

C-SimpleQA (Correct) 64.1

Research

DeepSeek R1
DeepSeek V3
DeepSeek Coder V2
DeepSeek VL
DeepSeek V2
DeepSeek Coder
DeepSeek Math
DeepSeek LLM

Product

DeepSeek App
DeepSeek Chat
DeepSeek Platform
API Pricing
Service Status

Legal & Safety

Privacy Policy
Terms of Use
Report Vulnerabilities

Join Us

Job Description

© 2025 DeepSeek. All rights reserved.

浙ICP备2023025841号

浙公⽹安备 33010502011812 号

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy