●Change8

Migrating to vLLM v0.10.0

Version v0.10.0 introduces 6 breaking changes. This guide details how to update your code.

Released: 7/24/2025

6
Breaking Changes
5
Migration Steps
10
Affected Symbols

⚠️ Check Your Code

If you use any of these symbols, you need to read this guide:

V0 enginePromptAdaptersPhi3-SmallBlockSparse AttentionSpec Decode workersLlamaForSequenceClassificationAutoWeightsLoaderget_tokenizer_infoFusedMoEModularKernelMultiModalHasher.hash_prompt_mm_data

Breaking Changes

●Issue #1

Removed V0 CPU/XPU/TPU/HPU backends. Users must migrate to V1 or supported backends.

●Issue #2

Removed long context LoRA support.

●Issue #3

Removed Prompt Adapters.

●Issue #4

Removed Phi3-Small & BlockSparse Attention support.

●Issue #5

Removed Spec Decode workers.

●Issue #6

Default model changed to Qwen3-0.6B from previous default.

Migration Steps

  1. 1
    Update PyTorch to 2.7.1 for CUDA environments.
  2. 2
    Update FlashInfer to v0.2.8rc1.
  3. 3
    If using CPU/XPU/TPU/HPU, ensure compatibility with the V1 engine as V0 backends are removed.
  4. 4
    Review model configurations if using Phi3-Small or Prompt Adapters as they are no longer supported.
  5. 5
    Update CLI scripts if relying on the previous default model (now Qwen3-0.6B).

Release Summary

v0.10.0 introduces the V1 engine as the primary focus, removing several legacy V0 backends and features while adding support for Llama 4 and NVIDIA Blackwell optimizations. It features significant performance improvements via async scheduling and microbatch tokenization.

Need More Details?

View the full release notes and all changes for vLLM v0.10.0.

View Full Changelog