vLLM 0.28.0 landed with 584 commits, delivering stack-wide Kimi-K3 optimizations, DeepSeek-V4 sparse MLA support, adaptive speculative decoding with 65% TTFT gains, and Model Runner V2 disaggregation.

Key Takeaways

  • ✓Adaptive speculative decoding with DFlash2/DSpark cuts time-to-first-token by 55-65% across deployments;
  • ✓Native full-stack acceleration for cutting-edge architectures including Kimi-K3 and DeepSeek-V4 sparse MLA;
  • ✓Broad hardware enablement across NVIDIA SM12x, AMD ROCm fused kernels, and Intel XPU backends.
⚡

Key Decision Metrics at a Glance

TL;DR Verdict
🏆 Benchmark RankRank
Adaptive speculative decoding with DFlash2/DSpark cuts time-to-first-token by 55-65% across deployments;
Full Leaderboard→
💰 Cost & PlansCompare
Kimi K3 · in: $3 / out: $15
Compare Plans↗
🛠️ Ecosystem & IDERules
Cursor · Claude Code · API
Setup IDE Rules↗
🧭

Not enough VRAM? Compare cloud API and self-hosting costs

Compare 40 dev plans & simulate token costs vs $20/mo subscriptions

⚔️ Compare Model→💰 Cloud cost calculator↗
Action HubReady to adopt this in production?

Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.

⚔️ Launch VS Compare→💳 View Plans↗
Up NextMiniMax M3.1 Flash weights now open on Hugging Face: 428B-total / 23B-active MoE, $0.15 input and $0.60 output per million tokens, 73.7% on DeepSWE
Read Now→