vLLM 0.34 Stops Reserving 8% of Your GPU: Extensible KV Cache, the Block Size 64 Default, and What to Delete From Your Launch Flags
TL;DR vLLM v0.34.0 (GitHub release 2026-11-16) changes three defaults that touch every vllm serve command line: the KV cache is sized from memory measured after warmup and CUDA graph capture, gpu_memory_utilization defaults to 1.0 when you do not set it, and --block-size defaults to 64 instead of 1…
Read the full story at DEV Community — AI ↗
Timeline · 1 report
- 2026-11-18 14:04 · DEV Community — AI
vLLM 0.34 Stops Reserving 8% of Your GPU: Extensible KV Cache, the Block Size 64 Default, and What to Delete From Your Launch Flags
More stories
- OpenAI launches GPT-6.2 Astra and Sol as Google ships Gemini 4 Neon and Microsoft unveils MAI Code 2 — OpenAI News
- Anthropic debuts on Nasdaq with record $81.3 billion IPO, hitting $2 trillion value — Anthropic (X)
- Google launches Gemini 4 Neon and opens Genie 4 world model to Gemini API developers — Google DeepMind YouTube
- Google books Blue Origin's New Glenn for 2029 launch of AI server satellites — Google: The Keyword
- AWS expands SageMaker AI with data preparation tools and cross-Region inference routing — AWS Machine Learning Blog
- Anthropic's stock debut soars, but Wall Street indexes drop — Washington Post AI
- AWS expands SageMaker JumpStart with OpenTSLM, model leaderboard, and PolyAI Dialog-RSN-1 — AWS Machine Learning Blog
- How Duco transforms financial data automation with Amazon Bedrock AgentCore — AWS Machine Learning Blog
Read the Daily Brief: the day's most important AI stories →