fix(compile_utils, ep_moe): update environment variable and dtype check (#12034)

This commit is contained in:
ishandhanani
2025-10-24 11:00:12 -07:00
committed by GitHub
parent d7f0d88fa2
commit 14203432b4
3 changed files with 6 additions and 5 deletions
+1 -1
View File
@@ -36,7 +36,7 @@ SGLang supports various environment variables that can be used to configure its
| `SGLANG_JIT_DEEPGEMM_PRECOMPILE` | Enable precompilation of DeepGEMM kernels | `"true"` |
| `SGLANG_JIT_DEEPGEMM_COMPILE_WORKERS` | Number of workers for parallel DeepGEMM kernel compilation | `4` |
| `SGL_IN_DEEPGEMM_PRECOMPILE_STAGE` | Indicator flag used during the DeepGEMM precompile script | `"false"` |
| `SGL_DG_CACHE_DIR` | Directory for caching compiled DeepGEMM kernels | `~/.cache/deep_gemm` |
| `SGLANG_DG_CACHE_DIR` | Directory for caching compiled DeepGEMM kernels | `~/.cache/deep_gemm` |
| `SGL_DG_USE_NVRTC` | Use NVRTC (instead of Triton) for JIT compilation (Experimental) | `"0"` |
| `SGL_USE_DEEPGEMM_BMM` | Use DeepGEMM for Batched Matrix Multiplication (BMM) operations | `"false"` |