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

[diffusion] fix: further align ltx2.3 accuracy with tp (#24660)

Browse Source
This commit is contained in:
Mick
2026-05-11 13:42:08 +08:00
committed by GitHub
parent ed70226ec1
commit 6e5b4de01a
16 changed files with 435 additions and 311 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs_new/cookbook/diffusion/LTX/LTX2 & LTX2.3.mdx
+7 -6
View File
@@ -71,7 +71,7 @@ Other deployment flags:
- `--lora-weight-name`: Select the exact safetensors file when the LoRA repository contains multiple weight files.
<Note>
For native LTX-2.3 two-stage serving without a user LoRA, `resident` is the fastest high-VRAM path. When you pass `--lora-path`, SGLang still applies the user LoRA during the two-stage switch, so use `resident` on H200-class GPUs for enough VRAM, but do not expect the same premerged-stage2 benefit as the no-user-LoRA path.
For native LTX-2.3 two-stage serving without a user LoRA, `resident` is the fastest high-VRAM path. LTX-2 still applies the distilled LoRA during the stage switch, so `--ltx2-two-stage-device-mode` is mainly an LTX-2.3 optimization. When you pass `--lora-path`, SGLang still applies the user LoRA during the two-stage switch, so use `resident` on H200-class GPUs for enough VRAM, but do not expect the same premerged-stage2 benefit as the no-user-LoRA path.
</Note>
### 3.3 Fast multi-GPU presets
@@ -80,11 +80,12 @@ For latency-oriented LTX serving, prefer CFG parallel over sequence parallelism.
| Target | Recommended server flags | Notes |
| --- | --- | --- |
| 1 high-VRAM GPU | `--ltx2-two-stage-device-mode resident` | Fastest two-stage setup when both DiTs fit. |
| 1 standard GPU | `--ltx2-two-stage-device-mode snapshot` | Lower VRAM than `resident`; use this when H100-class memory is tight. |
| 2 GPUs | `--num-gpus 2 --enable-cfg-parallel --ltx2-two-stage-device-mode resident` | Fastest common 2-GPU setup. |
| 4 GPUs | `--num-gpus 4 --tp-size 2 --enable-cfg-parallel --ltx2-two-stage-device-mode resident` | Fastest common 4-GPU layout: TP2 inside each CFG branch. |
| Official comparison | `--ltx2-two-stage-device-mode original` | Use this only when matching the original stage-switch semantics matters. |
| LTX-2.3, 1 high-VRAM GPU | `--ltx2-two-stage-device-mode resident` | Fastest two-stage setup when both DiTs fit. |
| LTX-2.3, 1 standard GPU | `--ltx2-two-stage-device-mode snapshot` | Lower VRAM than `resident`; use this when H100-class memory is tight. |
| LTX-2, 2 GPUs | `--num-gpus 2 --enable-cfg-parallel` | Fastest verified 2-GPU setup; keep `--dit-layerwise-offload` disabled unless memory is tight. |
| LTX-2.3, 2 GPUs | `--num-gpus 2 --enable-cfg-parallel --ltx2-two-stage-device-mode resident` | Fastest common 2-GPU setup. |
| LTX-2.3, 4 GPUs | `--num-gpus 4 --tp-size 2 --enable-cfg-parallel --ltx2-two-stage-device-mode resident` | Fastest common 4-GPU layout: TP2 inside each CFG branch. |
| Official comparison | `--ltx2-two-stage-device-mode original` | Use this only when matching the original LTX-2.3 stage-switch semantics matters. |
Use `--enable-cfg-parallel` for degree-2 CFG parallel. Use `--cfg-parallel-size` only when you explicitly need a different CFG branch count. If `resident` exceeds available VRAM, keep the same parallelism preset and switch only the device mode to `snapshot`.