DeepSeek Open-Sources deepseek-harness & DeepSelect: DSA Sparse Attention and DeepGEMM for High-Throughput Inference
DeepSeek has published deepseek-harness, the inference and plugin orchestration harness powering its web and API infrastructure.
DeepSelect & DeepGEMM
deepseek-harness open-sources DeepSelect—a GPU-accelerated routing kernel that evaluates attention sparsity in real-time, executing Dynamic Sparse Attention (DSA) with near-zero latency overhead. Paired with DeepGEMM FP8 matrix operations, self-hosted clusters achieve up to a 3.4× boost in tokens-per-second-per-GPU.
Source & Fact Check
This technical dispatch was verified against primary documentation released by DeepSeek GitHub.
Read Original Announcement on DeepSeek GitHub →Go deeper: related guide
This story makes more sense when you build it. Here is the practical step-by-step guide:
Claude Code Complete Guide: build your first autonomous coding agent →Get 1 AI agent blueprint every week
Free newsletter: one practical AI agent or automation you can build and sell, every week. No spam.