[Frontend] Migrate Responses harmony input validation to VLLMValidationError (#55701)

Adababy committed about 2 months ago

[Bugfix][Spec Decode] Resolve n_predict from text_config for Qwen3.5 multimodal MTP (#55369)

Soumyajit Ghosh committed about 2 months ago

[Frontend] Add stateless /v1/responses/render endpoint (#50195)

Francisco Javier Arceo committed about 2 months ago

[Bugfix][Quantization][XPU] Fix moe_wna16 linear weight loading (#52651)

Artur Fierka committed about 2 months ago

[Bugfix] OffloadingConnector: stop zeroing offload hits under MTP/EAGLE spec decode (#52771)

Kam Basra committed about 2 months ago

[Qwen3.8-Flash-Next] Support FP8 indexer cache for QSA (#54890)

Thien Tran committed about 2 months ago

[Docs] Add OLMo 2 to batch-invariance tested models (#55691)

Sean Westfall committed about 2 months ago

[XPU][LoRA] Support LoRA for DeepSeek V4 on XPU (#53689)

Chaojun Zhang committed about 2 months ago

[XPU] Route grouped_topk to the fused _moe_C kernel on XPU (#53580)

Marceli Fylcek committed about 2 months ago

[Frontend] Migrate Responses API validation errors to VLLMValidationError (#50257)

Adababy committed about 2 months ago

[ROCm][CI] Add attention-sink support to ROCm AITER sparse MLA (#54404)

Andreas Karatzas committed about 2 months ago

[CI] [Test] skip test_wna16_cuda_high_bit_skips_humming on non-CUDA platforms (#55660)

Chaojun Zhang committed about 2 months ago

[KVConnector] Guard lmcache_mp_connector state transition with num_external_tokens (#47505)

AlexHuang committed about 2 months ago

[Bugfix] DSv4 MXFP4 selector: stop narrowing explicit aliases to their BF16 variant (#53586)

Gabriel Wu committed about 2 months ago

[Bugfix][MooncakeStore] Fix finish-time save crash on hybrid models (#54643)

Zhewen Li committed about 2 months ago

[CI][ROCm] Temporarily skip unsupported HY-V4 initialization (#55653)

Andreas Karatzas committed about 2 months ago

[EPD] Add ECMooncakeConnector for encoder cache over Mooncake TransferEngine (#41567)

Teng Ma committed about 2 months ago

[Bugfix][Audio] Restore soundfile-first automatic decoding (#55642)

Andreas Karatzas committed about 2 months ago

[XPU] Use fused_input_norm kernel in FusedInputNorm (#52945)

zofia committed about 2 months ago

[Qwen3.8-Flash-Next] Remove torch.compile for NVIDIA implementation (#55272)

Thien Tran committed about 2 months ago