[Model Runner V2] Minor refactoring for EncoderRunner (#35628)

Woosuk Kwon committed 8 months ago

[AMD][CI] Support Triton attention with ExampleConnector (#34931)

Ryan Rock committed 8 months ago

Fix typo: implictly -> implicitly in isaac.py docstring (#35646)

lin-shh committed 8 months ago

[Bugfix][Model] Fix Qwen3.5/Qwen3Next ignoring --dtype flag on older GPUs (#35617)

lailoo committed 8 months ago

Add TMA support to fused_moe_lora kernel (#32195)

gnovack committed 8 months ago

[Model Runner V2] Add ModelStateInterface [4/N] (#35621)

Woosuk Kwon committed 8 months ago

[Bugfix] Fix Anthropic API base64 image handling in Messages endpoint (#35557)

Martin Vit committed 8 months ago

[Chore] Cleanup BNB utilization dead code (#35620)

Isotr0py committed 8 months ago

[Deprecation] Deprecate code in 0.17 as scheduled (#35441)

Wentao Ye committed 8 months ago

[Benchmark] Avoid unnecessary video download in MMVU (#35618)

Cyrus Leung committed 8 months ago

[Fix] Avoid sending image input to other PP ranks (#35405)

emricksini-h committed 8 months ago

Fix Qwen3_5MTP packed_modules_mapping for gate_up_proj (#35581)

cwazai committed 8 months ago

custom dataset img support base64 (#35280)

flutist committed 8 months ago

[Feat] Add CUDA torch fallbacks for fp8_mqa_logits/fp8_paged_mqa_logits_torch function (#35271)

Chauncey committed 8 months ago

add io_process_plugin for sparse embedding (#34214)

Augusto Yao committed 8 months ago

[Feature]Supports Anthropic Thinking Block (#33671)

Mario Hong committed 8 months ago

Add padding support to wvSplitK solution for skinny GEMMs (#33762)

Hashem Hashemi committed 8 months ago

[ROCm][CI] Parametrize vision score tests across attention backends with per-backend tolerances (#35571)

Andreas Karatzas committed 8 months ago

[Benchmark] Improve UX of sweep scripts (#35600)

Cyrus Leung committed 8 months ago

[Benchmark] Rename SLA Finder to Workload Explorer (#35586)

Cyrus Leung committed 8 months ago