What happened: SGLang and Miles Add Day-0 Support for DeepSeek-V4.1

SGLang and Miles added day-zero support for DeepSeek-V4.1, implementing optimizations for sliding-window attention, compressed KV caching, and Engram memory retrieval.

Why it matters: SGLang and Miles Add Day-0 Support for DeepSeek-V4.1

Engineers deploying DeepSeek-V4.1 need this to serve the model efficiently with reduced memory overhead and faster inference throughput.

Watch out: SGLang and Miles Add Day-0 Support for DeepSeek-V4.1

Bounded replay is an approximation that truncates local attention; decoder-side tail-only computation does not support input logprobs or full prompt hidden-state capture.

In the news

SGLang and Miles Add Day-0 Support for DeepSeek-V4.1

LMSYS · Published September 10, 2026 · 3 min read

In 30 seconds

What happened: SGLang and Miles added day-zero support for DeepSeek-V4.1, implementing optimizations for sliding-window attention, compressed KV caching, and Engram memory retrieval.
Why it matters: Engineers deploying DeepSeek-V4.1 need this to serve the model efficiently with reduced memory overhead and faster inference throughput.
Watch out: Bounded replay is an approximation that truncates local attention; decoder-side tail-only computation does not support input logprobs or full prompt hidden-state capture.

Read the full article

deepseek

The patterns behind this

Each one covers how the technique works, when it earns its cost, and where it breaks.

The Agent Architect

One pattern, one tradeoff, one production failure story. A short weekly briefing for people building agentic systems.

Weekly email, one-click unsubscribe. We only use your address to send the briefing.