minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

Revert "[Feature] NVFP4 Marlin fallback for non-Blackwell GPUs (SM75+… (#22047)

Browse Source
This commit is contained in:
Brayden Zhong
2026-04-03 13:12:30 -07:00
committed by GitHub
parent 0c9dc098e7
commit 6aafe756b9
16 changed files with 95 additions and 1410 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/references/environment_variables.md
-2
View File
@@ -121,8 +121,6 @@ SGLang supports various environment variables that can be used to configure its
| `SGLANG_INT4_WEIGHT` | Enable INT4 weight quantization | `false` |
| `SGLANG_PER_TOKEN_GROUP_QUANT_8BIT_V2` | Apply per token group quantization kernel with fused silu and mul and masked m | `false` |
| `SGLANG_FORCE_FP8_MARLIN` | Force using FP8 MARLIN kernels even if other FP8 kernels are available | `false` |
| `SGLANG_FORCE_NVFP4_MARLIN` | Force using NVFP4 Marlin fallback kernels even on Blackwell GPUs with native FP4 support | `false` |
| `SGLANG_FLASHINFER_FP4_GEMM_BACKEND` (deprecated) | Select backend for `mm_fp4` on Blackwell GPUs. **DEPRECATED**: Please use `--fp4-gemm-backend` instead. | `` |
| `SGLANG_NVFP4_CKPT_FP8_GEMM_IN_ATTN` | Quantize q_b_proj from BF16 to FP8 when launching DeepSeek NVFP4 checkpoint | `false` |
| `SGLANG_MOE_NVFP4_DISPATCH` | Use nvfp4 for moe dispatch (on flashinfer_cutlass or flashinfer_cutedsl moe runner backend) | `"false"` |
| `SGLANG_NVFP4_CKPT_FP8_NEXTN_MOE` | Quantize moe of nextn layer from BF16 to FP8 when launching DeepSeek NVFP4 checkpoint | `false` |