DeepSeek
Explained projects
deepseek-harness: DeepSeek Harness: The Agent Framework That Makes Itself Keep the Receipts
dsh turns documentation, translation, and architectural decisions into enforced runtime discipline, then wraps it in a plugin system built for modular agent infrastructure.
DeepSpec: The Full-Stack Machine for Teaching Small Models to Predict Big Ones
How DeepSeek-AI turned speculative decoding into a trainable, measurable pipeline, and why the real price of speed is often paid in storage, not compute.
Yet to be explained
DeepSeek-OCR
Explain
Contexts Optical Compression
3FS
Explain
A high-performance distributed file system designed to address the challenges of AI training and inference workloads.
Engram
Explain
Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models
DeepSeek-OCR-2
Explain
Visual Causal Flow
DualPipe
Explain
A bidirectional pipeline parallelism algorithm for computation-communication overlap in DeepSeek V3/R1 training.
TileKernels
Explain
A kernel library written in tilelang
profile-data
Explain
Analyze computation-communication overlap in V3/R1.
LPLB
Explain
An early research stage expert-parallel load balancer for MoE models based on linear programming.
DeepSelect
Explain
DeepSelect: TopK kernels for DeepSeek Sparse Attention (DSA) and Samplers