[Rust Frontend] Add support for TLS in render server (#54999)

Wen Zhou committed about 2 months ago

[CI] Zen5 image build (#50314)

Andy Linfoot committed about 2 months ago

[CI] Avoid logging test server environment values (#54379)

Taneem Ibrahim committed about 2 months ago

[ROCm][Perf] Optimize MiniMax-M3 decode indexer and top-k (#54682)

Fangzhou Ai committed about 2 months ago

[warmup] overlap renderer warmup and engine core initialization (#54557)

Ning Xie committed about 2 months ago

Fix DeepSeek V4 FlashMLA auto KV cache dtype (#45091)

elehayym committed about 2 months ago

[Rust Frontend] Support `--lora-modules` for static adapter loading (#54837)

Will Eaton committed about 2 months ago

[Rust Frontend] Use token-attributed text in reasoning and unified parsers (#54884)

Bugen Zhao committed about 2 months ago

[ROCm][Perf] Add low-M FP32 router GEMM for gfx950 (#54845)

Fangzhou Ai committed about 2 months ago

[Fusion] Manual `ActivationQuantFusionPass` initial application (#51415)

Misha Goin committed about 2 months ago

[Bugfix][Model] Fix FP8 PLE loading in mixed ModelOpt checkpoints (#54882)

sychen52 committed about 2 months ago

[Kernel] Enable Kimi-K3 SiTU on the CuteDSL MoE backend and the SM107 low-latency GEMM plan (#54606)

Bolin Sun committed about 2 months ago

[Model] add GLM-5.3-Flash support (#53906)

Jiangyun Zhu committed about 2 months ago

[Bugfix][KV Offload] Do not let a recurrent group's unhashed block truncate the load boundary (#52807)

Yifan Jiang committed about 2 months ago

[Bugfix][KV Connector] Populate SimpleCPUOffload BlockStored metadata (#54325)

mevince committed about 2 months ago

[Core] Deprecate "all" mamba cache mode (#55041)

Nick Hill committed about 2 months ago

[Bugfix][KV Connector] Safely fill circular buffers in DecodeBench (#54879)

Jeff (Junze) Ma committed about 2 months ago

Optimize PLE MTP metadata transfers (#55054)

bhsueh_NV committed about 2 months ago

[Core] Triton kernel for small-batch top-p only masking (#54651)

Nick Hill committed about 2 months ago

feat: Add support for reasoning_token_count to reasoning parser (#54982)

JasonCohere committed about 2 months ago