vLLM ships Day-0 support for Step 5 Preview: sparse attention, MoE, MTP and vision
vllm_project · x · 2026-10-15
vLLM announced Day-0 support for Step 5 Preview, spanning sparse attention, MoE, multi-token prediction and vision, with serving work across both Blackwell and Hopper GPUs.
Key details:
- Sparse attention: history is summarized into region summaries; an indexer scores and selects regions, then attention runs only over selected tokens, with unfinished regions kept in the local sliding window
- Backbone: combines sliding-window and sparse attention with MoE and MTP, alongside a vision processing path
- Engineering: work covers the sparse indexer, FP8 MegaMoE and cache transfer; follow-up posts report kernel benchmarks (indexer logits kernel latency -43%, top-k kernel -84%, overall prefill -2.2%)
Related event: vLLM Ships Day-0 Support for Step 5 Preview Across Blackwell and Hopper(7 posts)→
More from Infra
- Moving 1PB out of the cloud costs up to $74K in egress fees alone, $500K all-in — DavidLinthicum · 2026-10-15
- Robotics data gap: just 300K hours of manipulation data globally vs 100M+ needed, $3B spend projected — 0xSammy · 2026-10-15
- Ramp raises $1.85B at $60B valuation, says 85,000 firms save $1B+/month with AI tools — SethGRosenberg · 2026-10-15
- AI's smartest bear case is supply: $500B GPUs idle, but demand is elastic — bittingthembits · 2026-10-15
- Intel said to have secured foundry commitments from Broadcom and Google, Apple wins on 14A — BenBajarin · 2026-10-15
- Magnitude open-source inference engine beats Ollama by over 2x with auto-tuned kernels — nickbaumann_ · 2026-10-15