[KV Connector][Offloading] Disable parallel-agnostic fs-tier cache on V2 model runner (#46044)

Itay Etelis committed 4 months ago

[Tests] Add Qwen3 streaming parser delta boundary cases (#45708)

Palaiologos1453 committed 4 months ago

[ROCm][Spec Decode] Fix probabilistic draft probs test attention backend (#45706)

stefankoncarevic committed 4 months ago

[CI]Enable mxfp4 lora test for ROCm platform (#43802)

qli88 committed 4 months ago

[Bugfix][Kernel] Check output alignment in vectorize_with_alignment (fixes misaligned-address crash for non-multiple-of-8 head sizes) (#45466)

Humphrey committed 4 months ago

[ROCm][CI] move lora%N test to mi300 and gate (#45970)

Divakar Verma committed 4 months ago

[Frontend] Add Streaming Parser Engine and new GLM4.7/GLM5.1/GLM5.2 Parser (#45915)

Chauncey committed 4 months ago

[Bugfix][ROCm] Fix rocm_aiter_per_tensor_quant custom op aliasing (#45747)

Rohan Potdar committed 4 months ago

[Bugfix] Restore is_sym guard for zp in GPTQ/CT MoE to fix symmetric quant regression (#45656)

Yuwen Zhou committed 4 months ago

[Bugfix][Tool Parser] Handle non-finite numbers in coerce_to_schema_type (#43984)

Ashish Patel committed 4 months ago

Temporarily remove @markmc from CODEOWNERS (#46053)

Mark McLoughlin committed 4 months ago

[Model] Remove BambaForCausalLM (#45990)

Tiezhen WANG committed 4 months ago

Fix _riscv_supports_rvv_vlen128() to detect RVV on hardware without zvl flags (#43179)

lyd1992 committed 4 months ago

[Rust Frontend]: Add `/get_world_size` route with static parallel size (#44801)

Ashar committed 4 months ago

[Kernel] Add PDL support for DeepGEMM kernel (#46006)

Jee Jee Li committed 4 months ago

[ROCm][DSv4] Functional fixes for DeepSeek V4 on MI300X/MI325X (#45681)

Tuukka Sarvi committed 4 months ago

fix(quantization): Fix AWQ dequantize on Intel XPU and refactor AutoAWQ config (#42727)

AlexHuang committed 4 months ago

[CPU] Skip Triton kernel monkey-patches when Triton-CPU is available (#44991)

Jonathan Mamou committed 4 months ago

[Rust Frontend] Return model metadata fields in /v1/models (#45950)

Tahsin Tunan committed 4 months ago

Fix Stale Encoder Cache After Weight Update (#45093)

littlecircle0730 committed 4 months ago