docs: add XL (4B DiT) model documentation#1005
Merged
Conversation
Update English documentation to cover the new XL (4B) DiT models (acestep-v15-xl-base, acestep-v15-xl-sft, acestep-v15-xl-turbo): - README.md: Add XL models to Model Zoo table, update GPU recommendation with DiT column and XL VRAM guidance - INSTALL.md: Add XL download commands, available models table entries, and update "Which Model Should I Choose?" with XL recommendations - GPU_COMPATIBILITY.md: Add "XL (4B) DiT" column to tier table showing support level per tier (❌/⚠️ /✅) - Tutorial.md: Add XL model section and expand DiT selection summary with XL variants Key facts documented: - XL weights ~9GB (bf16) vs ~4.7GB for 2B - Min VRAM: 16GB with offload, 20GB+ recommended - All LM models (0.6B/1.7B/4B) fully compatible with XL Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Contributor
📝 WalkthroughWalkthroughAdds XL (4B) DiT documentation across multiple locales: updates README and INSTALL with XL model entries and download commands, revises GPU compatibility/VRAM recommendation tables to include XL DiT support markers and offload guidance, and extends Tutorial pages with an “XL (4B) Models” section and DiT selection updates. Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 inconclusive)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@docs/en/INSTALL.md`:
- Around line 675-677: Replace the non-descriptive repeated "[Link]" labels for
the three model rows with clear, unique link text to satisfy MD059; update the
links for the entries "acestep-v15-xl-base", "acestep-v15-xl-sft", and
"acestep-v15-xl-turbo" so each uses a descriptive label (for example
"acestep-v15-xl-base on Hugging Face", "acestep-v15-xl-sft on Hugging Face",
"acestep-v15-xl-turbo on Hugging Face") instead of "[Link]".
In `@README.md`:
- Around line 256-258: The table rows use non-descriptive link text "[Link]" for
each model entry (`acestep-v15-xl-base`, `acestep-v15-xl-sft`,
`acestep-v15-xl-turbo`); update each markdown link so the anchor text describes
the destination (for example "Hugging Face — acestep-v15-xl-base", "Hugging Face
— acestep-v15-xl-sft", "Hugging Face — acestep-v15-xl-turbo" or similar) instead
of generic "[Link]" to satisfy MD059 and improve accessibility.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: fb349753-7d41-4285-b2b1-e9816e364b97
📒 Files selected for processing (4)
README.mddocs/en/GPU_COMPATIBILITY.mddocs/en/INSTALL.mddocs/en/Tutorial.md
docs/en/INSTALL.md
Outdated
Comment on lines
+675
to
+677
| | **acestep-v15-xl-base** | XL (4B) Base DiT — higher quality, ≥16GB VRAM | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-base) | | ||
| | **acestep-v15-xl-sft** | XL (4B) SFT DiT — higher quality, ≥16GB VRAM | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-sft) | | ||
| | **acestep-v15-xl-turbo** | XL (4B) Turbo DiT — higher quality, ≥16GB VRAM | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-turbo) | |
Contributor
There was a problem hiding this comment.
Use descriptive link labels to satisfy markdown lint (MD059).
At Line 675-677, repeated [Link] text is non-descriptive and triggers lint warnings.
Suggested doc-only patch
-| **acestep-v15-xl-base** | XL (4B) Base DiT — higher quality, ≥16GB VRAM | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-base) |
-| **acestep-v15-xl-sft** | XL (4B) SFT DiT — higher quality, ≥16GB VRAM | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-sft) |
-| **acestep-v15-xl-turbo** | XL (4B) Turbo DiT — higher quality, ≥16GB VRAM | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-turbo) |
+| **acestep-v15-xl-base** | XL (4B) Base DiT — higher quality, ≥16GB VRAM | [acestep-v15-xl-base on Hugging Face](https://huggingface.co/ACE-Step/acestep-v15-xl-base) |
+| **acestep-v15-xl-sft** | XL (4B) SFT DiT — higher quality, ≥16GB VRAM | [acestep-v15-xl-sft on Hugging Face](https://huggingface.co/ACE-Step/acestep-v15-xl-sft) |
+| **acestep-v15-xl-turbo** | XL (4B) Turbo DiT — higher quality, ≥16GB VRAM | [acestep-v15-xl-turbo on Hugging Face](https://huggingface.co/ACE-Step/acestep-v15-xl-turbo) |📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| | **acestep-v15-xl-base** | XL (4B) Base DiT — higher quality, ≥16GB VRAM | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-base) | | |
| | **acestep-v15-xl-sft** | XL (4B) SFT DiT — higher quality, ≥16GB VRAM | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-sft) | | |
| | **acestep-v15-xl-turbo** | XL (4B) Turbo DiT — higher quality, ≥16GB VRAM | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-turbo) | | |
| | **acestep-v15-xl-base** | XL (4B) Base DiT — higher quality, ≥16GB VRAM | [acestep-v15-xl-base on Hugging Face](https://huggingface.co/ACE-Step/acestep-v15-xl-base) | | |
| | **acestep-v15-xl-sft** | XL (4B) SFT DiT — higher quality, ≥16GB VRAM | [acestep-v15-xl-sft on Hugging Face](https://huggingface.co/ACE-Step/acestep-v15-xl-sft) | | |
| | **acestep-v15-xl-turbo** | XL (4B) Turbo DiT — higher quality, ≥16GB VRAM | [acestep-v15-xl-turbo on Hugging Face](https://huggingface.co/ACE-Step/acestep-v15-xl-turbo) | |
🧰 Tools
🪛 markdownlint-cli2 (0.22.0)
[warning] 675-675: Link text should be descriptive
(MD059, descriptive-link-text)
[warning] 676-676: Link text should be descriptive
(MD059, descriptive-link-text)
[warning] 677-677: Link text should be descriptive
(MD059, descriptive-link-text)
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@docs/en/INSTALL.md` around lines 675 - 677, Replace the non-descriptive
repeated "[Link]" labels for the three model rows with clear, unique link text
to satisfy MD059; update the links for the entries "acestep-v15-xl-base",
"acestep-v15-xl-sft", and "acestep-v15-xl-turbo" so each uses a descriptive
label (for example "acestep-v15-xl-base on Hugging Face", "acestep-v15-xl-sft on
Hugging Face", "acestep-v15-xl-turbo on Hugging Face") instead of "[Link]".
Comment on lines
+256
to
+258
| | `acestep-v15-xl-base` | ✅ | ❌ | ❌ | ✅ | 50 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | High | High | Easy | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-base) | | ||
| | `acestep-v15-xl-sft` | ✅ | ✅ | ❌ | ✅ | 50 | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | Very High | Medium | Easy | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-sft) | | ||
| | `acestep-v15-xl-turbo` | ✅ | ✅ | ❌ | ❌ | 8 | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | Very High | Medium | Medium | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-turbo) | |
Contributor
There was a problem hiding this comment.
Replace generic [Link] text with descriptive labels (MD059).
At Line 256-258, link text is non-descriptive and will keep markdownlint warnings active.
Suggested doc-only patch
-| `acestep-v15-xl-base` | ✅ | ❌ | ❌ | ✅ | 50 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | High | High | Easy | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-base) |
-| `acestep-v15-xl-sft` | ✅ | ✅ | ❌ | ✅ | 50 | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | Very High | Medium | Easy | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-sft) |
-| `acestep-v15-xl-turbo` | ✅ | ✅ | ❌ | ❌ | 8 | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | Very High | Medium | Medium | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-turbo) |
+| `acestep-v15-xl-base` | ✅ | ❌ | ❌ | ✅ | 50 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | High | High | Easy | [acestep-v15-xl-base on Hugging Face](https://huggingface.co/ACE-Step/acestep-v15-xl-base) |
+| `acestep-v15-xl-sft` | ✅ | ✅ | ❌ | ✅ | 50 | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | Very High | Medium | Easy | [acestep-v15-xl-sft on Hugging Face](https://huggingface.co/ACE-Step/acestep-v15-xl-sft) |
+| `acestep-v15-xl-turbo` | ✅ | ✅ | ❌ | ❌ | 8 | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | Very High | Medium | Medium | [acestep-v15-xl-turbo on Hugging Face](https://huggingface.co/ACE-Step/acestep-v15-xl-turbo) |📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| | `acestep-v15-xl-base` | ✅ | ❌ | ❌ | ✅ | 50 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | High | High | Easy | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-base) | | |
| | `acestep-v15-xl-sft` | ✅ | ✅ | ❌ | ✅ | 50 | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | Very High | Medium | Easy | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-sft) | | |
| | `acestep-v15-xl-turbo` | ✅ | ✅ | ❌ | ❌ | 8 | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | Very High | Medium | Medium | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-turbo) | | |
| | `acestep-v15-xl-base` | ✅ | ❌ | ❌ | ✅ | 50 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | High | High | Easy | [acestep-v15-xl-base on Hugging Face](https://huggingface.co/ACE-Step/acestep-v15-xl-base) | | |
| | `acestep-v15-xl-sft` | ✅ | ✅ | ❌ | ✅ | 50 | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | Very High | Medium | Easy | [acestep-v15-xl-sft on Hugging Face](https://huggingface.co/ACE-Step/acestep-v15-xl-sft) | | |
| | `acestep-v15-xl-turbo` | ✅ | ✅ | ❌ | ❌ | 8 | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | Very High | Medium | Medium | [acestep-v15-xl-turbo on Hugging Face](https://huggingface.co/ACE-Step/acestep-v15-xl-turbo) | |
🧰 Tools
🪛 markdownlint-cli2 (0.22.0)
[warning] 256-256: Link text should be descriptive
(MD059, descriptive-link-text)
[warning] 257-257: Link text should be descriptive
(MD059, descriptive-link-text)
[warning] 258-258: Link text should be descriptive
(MD059, descriptive-link-text)
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@README.md` around lines 256 - 258, The table rows use non-descriptive link
text "[Link]" for each model entry (`acestep-v15-xl-base`, `acestep-v15-xl-sft`,
`acestep-v15-xl-turbo`); update each markdown link so the anchor text describes
the destination (for example "Hugging Face — acestep-v15-xl-base", "Hugging Face
— acestep-v15-xl-sft", "Hugging Face — acestep-v15-xl-turbo" or similar) instead
of generic "[Link]" to satisfy MD059 and improve accessibility.
- Update min VRAM from ≥16GB to ≥12GB (offload makes 12GB viable) - Fix 16-20GB tier note: "XL requires CPU offload below 20GB" - GPU_COMPATIBILITY: clarify Tier 5 marginal note for 12-16GB range - Tutorial: use full model names (acestep-v15-xl-*) instead of short - INSTALL: fix comment dash style in download section Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Contributor
There was a problem hiding this comment.
♻️ Duplicate comments (2)
docs/en/INSTALL.md (1)
675-677:⚠️ Potential issue | 🟡 MinorReplace generic
[Link]labels with descriptive link text.These three rows still use non-descriptive link labels and will continue triggering markdownlint MD059; please switch to distinct labels per model.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/en/INSTALL.md` around lines 675 - 677, The three table rows for model entries (acestep-v15-xl-base, acestep-v15-xl-sft, and acestep-v15-xl-turbo) use the generic "[Link]" label which triggers markdownlint MD059; update each link to have unique, descriptive text (e.g., "acestep-v15-xl-base on Hugging Face", "acestep-v15-xl-sft on Hugging Face", "acestep-v15-xl-turbo on Hugging Face") so the anchor text is specific to the model and replaces the generic "[Link]" labels.README.md (1)
256-258:⚠️ Potential issue | 🟡 MinorUse descriptive Hugging Face link labels in the XL rows.
These anchors are still generic (
[Link]) and likely keep MD059 warnings active.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@README.md` around lines 256 - 258, The table rows for the XL models (`acestep-v15-xl-base`, `acestep-v15-xl-sft`, `acestep-v15-xl-turbo`) use generic `[Link]` anchors which trigger MD059; replace each `[Link]` with a descriptive Hugging Face label (e.g., `[acestep-v15-xl-base on Hugging Face]`, `[acestep-v15-xl-sft on Hugging Face]`, `[acestep-v15-xl-turbo on Hugging Face]`) so the anchor text is unique and descriptive while keeping the existing URLs unchanged.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Duplicate comments:
In `@docs/en/INSTALL.md`:
- Around line 675-677: The three table rows for model entries
(acestep-v15-xl-base, acestep-v15-xl-sft, and acestep-v15-xl-turbo) use the
generic "[Link]" label which triggers markdownlint MD059; update each link to
have unique, descriptive text (e.g., "acestep-v15-xl-base on Hugging Face",
"acestep-v15-xl-sft on Hugging Face", "acestep-v15-xl-turbo on Hugging Face") so
the anchor text is specific to the model and replaces the generic "[Link]"
labels.
In `@README.md`:
- Around line 256-258: The table rows for the XL models (`acestep-v15-xl-base`,
`acestep-v15-xl-sft`, `acestep-v15-xl-turbo`) use generic `[Link]` anchors which
trigger MD059; replace each `[Link]` with a descriptive Hugging Face label
(e.g., `[acestep-v15-xl-base on Hugging Face]`, `[acestep-v15-xl-sft on Hugging
Face]`, `[acestep-v15-xl-turbo on Hugging Face]`) so the anchor text is unique
and descriptive while keeping the existing URLs unchanged.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: fe082e31-22e5-4708-827b-d27e30f8f347
📒 Files selected for processing (4)
README.mddocs/en/GPU_COMPATIBILITY.mddocs/en/INSTALL.mddocs/en/Tutorial.md
✅ Files skipped from review due to trivial changes (1)
- docs/en/GPU_COMPATIBILITY.md
🚧 Files skipped from review as they are similar to previous changes (1)
- docs/en/Tutorial.md
Update translated documentation with XL model support: - zh: INSTALL.md (model recommendation table), GPU_COMPATIBILITY.md (tier table with XL column), Tutorial.md (XL section + summary) - ja: INSTALL.md, GPU_COMPATIBILITY.md, Tutorial.md (same updates) - ko: GPU_COMPATIBILITY.md, Tutorial.md (XL section + summary table added) All translations consistent with English docs: ≥12GB VRAM with offload, 20GB+ recommended, all LM models compatible with XL. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@docs/ko/GPU_COMPATIBILITY.md`:
- Around line 11-15: Update the batch-size cells for the listed GPU tiers in the
table so they match the configured limits: change 티어 3 batch value from "1 / 2"
to "2 / 2", change 티어 5 batch value from "2 / 4" to "4 / 4", and change 티어 6b
batch value from "4 / 8" to "8 / 8"; verify the same corrections are applied
consistently across other locale docs that show these tiers (refer to the rows
labeled "티어 3", "티어 5", and "티어 6b") to avoid desynchronization.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: a41786ee-2f58-4e9d-9690-152334183892
📒 Files selected for processing (8)
docs/ja/GPU_COMPATIBILITY.mddocs/ja/INSTALL.mddocs/ja/Tutorial.mddocs/ko/GPU_COMPATIBILITY.mddocs/ko/Tutorial.mddocs/zh/GPU_COMPATIBILITY.mddocs/zh/INSTALL.mddocs/zh/Tutorial.md
✅ Files skipped from review due to trivial changes (3)
- docs/zh/Tutorial.md
- docs/ko/Tutorial.md
- docs/ja/Tutorial.md
- ≥24GB tier: note xl-base for extract/lego/complete (en/zh/ja) - Fix pre-existing Korean GPU tier batch size discrepancies: Tier 3: 1/2 → 2/2, Tier 5: 2/4 → 4/4, Tier 6b: 4/8 → 8/8 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Contributor
There was a problem hiding this comment.
🧹 Nitpick comments (1)
docs/ko/GPU_COMPATIBILITY.md (1)
16-18: Standardize number+unit spacing for readability consistency.On Line 16 and Line 18, consider using spaced units (
24 GB,9 GB,4.7 GB) instead of24GB,9GB,4.7GBto keep terminology formatting consistent across the doc.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/ko/GPU_COMPATIBILITY.md` around lines 16 - 18, Replace contiguous number+unit tokens with a space between number and unit for consistency: change occurrences like "≥24GB", "24GB", "9GB", "4.7GB" to "≥24 GB", "24 GB", "9 GB", "4.7 GB" (and similarly "0.6B, 1.7B, 4B" if you want consistent spacing to "0.6 B, 1.7 B, 4 B" only where it matches doc style). Update the "XL (4B) DiT 열" row and the explanatory sentence ("XL 모델 가중치 약 9GB (bf16), 2B는 약 4.7GB") to use spaced units ("XL (4 B) DiT 열" and "약 9 GB (bf16), 2 B는 약 4.7 GB") to standardize formatting across the document.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Nitpick comments:
In `@docs/ko/GPU_COMPATIBILITY.md`:
- Around line 16-18: Replace contiguous number+unit tokens with a space between
number and unit for consistency: change occurrences like "≥24GB", "24GB", "9GB",
"4.7GB" to "≥24 GB", "24 GB", "9 GB", "4.7 GB" (and similarly "0.6B, 1.7B, 4B"
if you want consistent spacing to "0.6 B, 1.7 B, 4 B" only where it matches doc
style). Update the "XL (4B) DiT 열" row and the explanatory sentence ("XL 모델 가중치
약 9GB (bf16), 2B는 약 4.7GB") to use spaced units ("XL (4 B) DiT 열" and "약 9 GB
(bf16), 2 B는 약 4.7 GB") to standardize formatting across the document.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 04772c88-2e3d-4551-88fb-232410129a01
📒 Files selected for processing (5)
README.mddocs/en/INSTALL.mddocs/ja/INSTALL.mddocs/ko/GPU_COMPATIBILITY.mddocs/zh/INSTALL.md
✅ Files skipped from review due to trivial changes (3)
- docs/en/INSTALL.md
- docs/zh/INSTALL.md
- README.md
🚧 Files skipped from review as they are similar to previous changes (1)
- docs/ja/INSTALL.md
3 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Key facts: XL weights ~9GB bf16, min 16GB VRAM (offload), 20GB+ recommended. All LM models fully compatible with XL.
Test plan
Closes #993
🤖 Generated with Claude Code
Summary by CodeRabbit