Add Inkling model support (#31681)

Co-authored-by: Chunan Zeng <zcnrex@gmail.com>
Co-authored-by: Ke Bao <ispobaoke@gmail.com>
Co-authored-by: Yanbin Jiang <jybsuper@gmail.com>
Co-authored-by: Yuhao Yang <47235274+yhyang201@users.noreply.github.com>
Co-authored-by: Qiaolin Yu <qiaolin.yu@radixark.ai>
Co-authored-by: Zhichen Zeng <zczeng@uw.edu>
Co-authored-by: Aurick Qiao <aurick@thinkingmachines.ai>
Co-authored-by: Joseph <jk@thinkingmachines.ai>

python/pyproject.toml

@@ -33,6 +33,7 @@ dependencies = [
   "flash-attn-4==4.0.0b15",
   "flashinfer_python[cu13]==0.6.14", # keep it aligned with jit-cache version in Dockerfile
   "gguf",
+  "helion==0.2.6",
   "humming-kernels[cu13]==0.1.10",
   "interegular",
   "IPython",
@@ -42,6 +43,7 @@ dependencies = [
   "modelscope",
   "msgspec",
   "ninja",
+  "numba==0.65.1",
   "numpy",
   "nvidia-cutlass-dsl[cu13]==4.5.2",
   "nvidia-mathdx==25.6.0",
@@ -228,5 +230,11 @@ target = "sglang.srt.grpc._core"
 path = "../rust/sglang-grpc/Cargo.toml"
 binding = "PyO3"
 
+[[tool.setuptools-rust.ext-modules]]
+target = "sglang.srt.multimodal._core"
+path = "../rust/sglang-mm/Cargo.toml"
+binding = "PyO3"
+debug = false
+
 [tool.kernels.dependencies]
 "kernels-community/sgl-flash-attn3" = 1