[diffusion] model: support LTX2.3 high quality pipeline (#23366)

This commit is contained in:
Mick
2026-04-24 14:18:20 +08:00
committed by GitHub
parent 734e1e2965
commit cd1fa7506a
19 changed files with 1508 additions and 426 deletions
@@ -285,7 +285,7 @@ Optimization columns are abbreviated to keep the matrix readable:
<td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>❌</td>
</tr>
<tr>
<td style={{padding: "9px 12px", fontWeight: 500, backgroundColor: "rgba(255,255,255,0.02)"}}>LTX-2 (one and two stages)</td>
<td style={{padding: "9px 12px", fontWeight: 500, backgroundColor: "rgba(255,255,255,0.02)"}}>LTX-2 (one/two-stage/TI2V)</td>
<td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}><code>Lightricks/LTX-2</code></td>
<td style={{padding: "9px 8px", backgroundColor: "rgba(255,255,255,0.02)"}}>768×512<br />1536×1024</td>
<td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>❌</td>
@@ -297,9 +297,9 @@ Optimization columns are abbreviated to keep the matrix readable:
<td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>❌</td>
</tr>
<tr>
<td style={{padding: "9px 12px", fontWeight: 500, backgroundColor: "rgba(255,255,255,0.02)"}}>LTX-2.3 (one and two stages)</td>
<td style={{padding: "9px 12px", fontWeight: 500, backgroundColor: "rgba(255,255,255,0.02)"}}>LTX-2.3 (one/two-stage/TI2V/HQ)</td>
<td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}><code>Lightricks/LTX-2.3</code></td>
<td style={{padding: "9px 8px", backgroundColor: "rgba(255,255,255,0.02)"}}>768×512<br />1536×1024</td>
<td style={{padding: "9px 8px", backgroundColor: "rgba(255,255,255,0.02)"}}>768×512<br />1536×1024<br />1920×1088 (HQ default)</td>
<td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>❌</td>
<td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.02)"}}>❌</td>
<td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>❌</td>
@@ -315,12 +315,18 @@ Optimization columns are abbreviated to keep the matrix readable:
1. Wan2.2 TI2V 5B has some quality issues when performing I2V generation. We are working on fixing this issue.
2. SageSLA is based on SpargeAttn. Install it first with `pip install git+https://github.com/thu-ml/SpargeAttn.git --no-build-isolation`
3. LTX-2 and LTX-2.3 two-stage generation uses `--pipeline-class-name LTX2TwoStagePipeline`. The spatial upsampler and distilled LoRA are auto-resolved from the model snapshot by default, and can still be overridden with `--spatial-upsampler-path` and `--distilled-lora-path`.
- For LTX models, the `Resolution` column uses output video `width×height` semantics, matching `sglang generate --width ... --height ...`.
4. LTX-2.3 two-stage also supports `--ltx2-two-stage-device-mode &#123;legacy,snapshot,resident&#125;`:
3. LTX pipeline selection:
- One-stage: `--pipeline-class-name LTX2Pipeline`
- Two-stage: `--pipeline-class-name LTX2TwoStagePipeline`
- Two-stage HQ: `--pipeline-class-name LTX2TwoStageHQPipeline` (HQ defaults to 1920×1088; you can still override `--width/--height`)
- LTX-2 and LTX-2.3 support both T2V and TI2V (`--image-path`) on one-stage and two-stage pipelines (including HQ).
- The spatial upsampler and distilled LoRA are auto-resolved from the model snapshot by default, and can still be overridden with `--spatial-upsampler-path` and `--distilled-lora-path`.
- For LTX models, the `Resolutions` column uses output video `width×height` semantics, matching `sglang generate --width ... --height ...`.
4. LTX-2 / LTX-2.3 two-stage also supports `--ltx2-two-stage-device-mode {original,snapshot,resident}`:
- `snapshot` is the default and recommended mode.
- `resident` usually provides the best latency/throughput but uses much more VRAM.
- `legacy` preserves the historical switching path for fallback/debug.
- `original` keeps official two-stage semantics without the premerged stage-2 transformer path.
- Example (one prior run): `original` `154.67s`, `snapshot` `114.05s`, `resident` `75.71s`; peak VRAM trend is `original < snapshot < resident`.
### Image Generation Models