AI Disruption
Subscribe
Sign in
Home
Podcast
Chat
Chip
Meta
Paper
Qwen
Agent
Robot
OpenAI
YouTube
AI Code
AI Video
AI Weekly
Elon Musk
AI Writing
AI Painting
AI Business
🎈 Guest Posts
AI Open Source
Machine Learning
Chinese Outbound
Foundation Model
Archive
About
Chinese Outbound
Latest
Top
Discussions
Manus Ignites the Intelligent Agent Reproduction Craze! DeepSeek Has Been Integrated
Manus ignites an intelligent agent revolution—sparking rapid open-source reproductions, DeepSeek integration, and fresh GAIA benchmark insights.
Mar 9
Â
•
Â
Meng Li
3
Share this post
AI Disruption
Manus Ignites the Intelligent Agent Reproduction Craze! DeepSeek Has Been Integrated
Copy link
Facebook
Email
Notes
More
DeepSeek R1 Technology Successfully Migrates to the Multimodal Domain, Fully Open Sourced
Discover Visual-RFT—an open-source breakthrough that extends DeepSeek-R1’s rule-based reinforcement learning to vision-language models for efficient…
Mar 4
Â
•
Â
Meng Li
4
Share this post
AI Disruption
DeepSeek R1 Technology Successfully Migrates to the Multimodal Domain, Fully Open Sourced
Copy link
Facebook
Email
Notes
More
DeepSeek's GRPO: Complete From-Scratch Implementation
Discover how to implement GRPO from scratch using Qwen2.5-1.5B-Instruct in this comprehensive distributed RL tutorial to boost model performance and…
Mar 2
Â
•
Â
Meng Li
6
Share this post
AI Disruption
DeepSeek's GRPO: Complete From-Scratch Implementation
Copy link
Facebook
Email
Notes
More
DeepSeek Unveils V3/R1 Inference System with 545% Cost-Profit Ratio and Full Transparency!
Discover DeepSeek's V3/R1 Inference System, with optimized batch scaling, load balancing, and full cost transparency for maximum performance and…
Mar 1
Â
•
Â
Meng Li
2
Share this post
AI Disruption
DeepSeek Unveils V3/R1 Inference System with 545% Cost-Profit Ratio and Full Transparency!
Copy link
Facebook
Email
Notes
More
DeepSeek Open Source: 3FS & Smallpond for Effortless PB-Level Data Processing
DeepSeek’s 3FS & Smallpond are revolutionizing AI data processing. With impressive speed, scalability, and performance, these open-source tools handle…
Feb 28
Â
•
Â
Meng Li
2
Share this post
AI Disruption
DeepSeek Open Source: 3FS & Smallpond for Effortless PB-Level Data Processing
Copy link
Facebook
Email
Notes
More
DeepSeek Release: Boost AI Performance with DualPipe, EPLB, and Profile-Data
Discover DeepSeek's latest releases, including DualPipe, EPLB, and profile-data, designed to enhance AI performance and optimize…
Feb 27
Â
•
Â
Meng Li
1
Share this post
AI Disruption
DeepSeek Release: Boost AI Performance with DualPipe, EPLB, and Profile-Data
Copy link
Facebook
Email
Notes
More
DeepSeek Releases DeepGEMM: 300 Lines of Code Accelerate V3 & R1, R2 Expected Before May
DeepSeek unveils DeepGEMM, an FP8 GEMM library accelerating V3/R1 performance with 300 lines of code. Expect R2 model release before May for enhanced AI…
Feb 26
Â
•
Â
Meng Li
2
Share this post
AI Disruption
DeepSeek Releases DeepGEMM: 300 Lines of Code Accelerate V3 & R1, R2 Expected Before May
Copy link
Facebook
Email
Notes
More
DeepSeek Releases MoE EP Communication Library DeepEP – Truly Open!
DeepSeek has released DeepEP, an EP communication library for MoE model training and inference. This open-source project improves performance with…
Feb 25
Â
•
Â
Meng Li
Share this post
AI Disruption
DeepSeek Releases MoE EP Communication Library DeepEP – Truly Open!
Copy link
Facebook
Email
Notes
More
DeepSeek Releases FlashMLA, Boosting H800 GPU Performance
DeepSeek launches FlashMLA, an efficient decoding kernel for Nvidia's H800 GPU, boosting AI task performance and lowering training costs with MLA and…
Feb 24
Â
•
Â
Meng Li
4
Share this post
AI Disruption
DeepSeek Releases FlashMLA, Boosting H800 GPU Performance
Copy link
Facebook
Email
Notes
More
MoBA Attention by Kimi Yang: DeepSeek NSA Collision & Code Release
Discover the MoBA attention mechanism, an advanced approach combining MoE and FlashAttention for efficient long-sequence processing in large language…
Feb 20
Â
•
Â
Meng Li
1
Share this post
AI Disruption
MoBA Attention by Kimi Yang: DeepSeek NSA Collision & Code Release
Copy link
Facebook
Email
Notes
More
DeepSeek's Liang Wenfeng Unveils NSA: A Game-Changing Attention Architecture
DeepSeek's NSA introduces a fast, hardware-aligned sparse attention mechanism for efficient long-context training and inference in large models.
Feb 18
Â
•
Â
Meng Li
6
Share this post
AI Disruption
DeepSeek's Liang Wenfeng Unveils NSA: A Game-Changing Attention Architecture
Copy link
Facebook
Email
Notes
More
DeepSeek Launches CODEI/O: Enhancing Large Model Inference with Thought Chains
DeepSeek's CODEI/O dataset enhances model reasoning by transforming code into natural language thought chains, improving performance across various…
Feb 17
Â
•
Â
Meng Li
1
Share this post
AI Disruption
DeepSeek Launches CODEI/O: Enhancing Large Model Inference with Thought Chains
Copy link
Facebook
Email
Notes
More
Share
Copy link
Facebook
Email
Notes
More
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts