[Model Runner V2][Spec Decode] Add Gemma4 MTP support (#43241)

Giancarlo Delfin committed 5 months ago

[XPU] skip unapplied UT in test_gpu_model_runner.py (#44289)

Yan Ma committed 5 months ago

[Bugfix] Fix Gemma4 MTP block_table batch_size mismatch under concurrent load (#43982)

Dima committed 5 months ago

Bump actions/stale from 10.1.1 to 10.2.0 (#35078)

dependabot[bot] committed 5 months ago

[Minor] Remove FlashInfer version check in topk_topp_sampler (#44442)

Woosuk Kwon committed 5 months ago

[Bug Fix][Model Runner V2][Spec Decode] Warmup & capture with different attention states for speculator prefill (#44253)

Giancarlo Delfin committed 5 months ago

[Bugfix] Honor tool_choice="none" in Chat Completions streaming (#42752)

hoobnn committed 5 months ago

[Feature] Support batch invariant rms norm with residual (#42453)

Wentao Ye committed 5 months ago

[Model] Add Gemma4 Unified (encoder-free) support (#44429)

Luciano Martins committed 5 months ago

[LoRA] Fix dedup for post-replacement module aliases (#44413)

linitra24 committed 5 months ago

[Refactor] Remove dead code fp quant (#44122)

Wentao Ye committed 5 months ago

[ROCm][CI] Move Model Executor test step from MI250 to MI300 (gfx942) (#44370)

JartX committed 5 months ago

[10b/n] Migrate custom all-reduce, DeepSeek V4 fused MLA, MiniMax reduce-RMS, and MXFP8 MoE to libtorch stable ABI (#44365)

Chris Leonard committed 5 months ago

Handle spinloop ext load failure gracefully (#43659)

pschlan-amd committed 5 months ago

fix(config): validate max_num_scheduled_tokens >= 0 on all paths (#44207)

Willow Lopez committed 5 months ago

[KVCache] Support Pluggable KVCacheSpec (#37505)

Mengqing Cao committed 5 months ago

[CI] Align PD tests to HMA on by default (#44174)

Nicolò Lucchesi committed 5 months ago

[CI/Build] Fix LoRA testing (#44425)

Jee Jee Li committed 5 months ago

[Model Runner V2] Use FlashInfer sampler (#42472)

Nick Hill committed 5 months ago

[Frontend][Responses API] Move developer-to-system conversion into HF renderer (#43590)

Chauncey committed 5 months ago