SolitaryThinkerandClaude Opus 5.5
d692d1cc3b
These ports were marked pending removal in #1820. Each has a newer, maintained counterpart in the repo, and their model-specific hooks leaked into shared stages, request fields, CLI args, and loaders. Remove each model's code, configs, presets, examples, converters, tests, cookbook pages, CI planner rules, and registry entries, along with the shared-code hooks only they used: - camera kwargs (GameCraft camera_states, DreamX y_camera) in the shared denoising loop - LongCat refine branches in latent preparation, the decoding crop, the --bsa-* CLI args, its train model, and third_party/longcat_video - Z-Image cfg_normalization/cfg_truncation and GameCraft/DreamX camera/action fields in SamplingParam, ForwardBatch, and the schema - the GLM-Image vision-language loader and the TurboDiffusion rCM scheduler and SLA auto-select in the train entrypoint - the Wan definition groups that only preserved DreamX detector order Removed env vars and their legacy aliases move to DEPRECATED_VARIABLES per the env-var policy. Matrix Game 2.0 stays: the new trainer's world model scenarios depend on it. The SLA and BSA attention backends stay. The TurboDiffusion int8 kernels in fastvideo-kernel are left for a separate kernel PR. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
21 KiB
Compatibility Matrix
This page summarizes common model + optimization combinations.
For the canonical, code-level list of model IDs recognized by
VideoGenerator.from_pretrained(...), see the registrations in
fastvideo/registry.py (register_configs(...) entries).
!!! note
The full STA integration in fastvideo/ is archived from main and kept
in sta_do_not_delete:
https://github.com/hao-ai-lab/FastVideo/tree/sta_do_not_delete
We do this because we believe VSA is strictly better than STA for the
actively maintained main inference path.
Registered Model IDs
Every Hugging Face model ID registered in fastvideo/registry.py on main
(commit 8d89f30d), grouped by family. Any ID below can
be passed to VideoGenerator.from_pretrained(...); FastVideo resolves the
matching pipeline and sampling defaults. The Family column is a
documentation grouping: it follows each registration's declared model_family,
except black-forest-labs/FLUX.1-dev, which declares none and is listed under
flux for readability. The Workloads column shows each
registration's declared workload_types; — means the entry is registered
without a UI workload option but is still loadable by ID. The Example
column links a runnable script in examples/inference/basic/ where one exists.
| Family | HuggingFace Model ID | Workloads | Example |
|---|---|---|---|
| cosmos | nvidia/Cosmos-Predict2-2B-Video2World |
T2V | — |
| cosmos25 | KyleShao/Cosmos-Predict2.5-2B-Diffusers |
T2V | basic_cosmos2_5_t2w.py |
| cosmos25 | nvidia/Cosmos-Predict2.5-14B |
T2V | basic_cosmos2_5_t2w.py |
| flux | black-forest-labs/FLUX.1-dev |
T2I | basic_flux_dev.py |
| flux2 | black-forest-labs/FLUX.2-klein-4Bblack-forest-labs/FLUX.2-klein-9B |
T2I | basic_flux2_klein.py |
| flux2 | black-forest-labs/FLUX.2-dev |
T2I | basic_flux2.py |
| gen3c | FastVideo/GEN3C-Cosmos-7B-Diffusers |
T2V | basic_gen3c.py |
| hunyuan | hunyuanvideo-community/HunyuanVideo |
T2V | — |
| hunyuan | FastVideo/FastHunyuan-diffusers |
T2V | — |
| hunyuan15 | hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v |
T2V | basic_hy15.py |
| hunyuan15 | hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_i2v_step_distilled |
I2V | basic_hy15.py |
| hunyuan15 | hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-720p_t2v |
T2V | basic_hy15.py |
| hunyuan15 | hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-720p_i2v_distilled |
I2V | basic_hy15.py |
| hunyuan15 | weizhou03/HunyuanVideo-1.5-Diffusers-1080pweizhou03/HunyuanVideo-1.5-Diffusers-1080p-2SR |
— | basic_hy15_1080p.py |
| hyworld | FastVideo/HY-WorldPlay-Bidirectional-Diffusers |
— | basic_hyworld.py |
| kandinsky5 | kandinskylab/Kandinsky-5.0-T2V-Lite-sft-5s-Diffusers |
T2V | basic_kandinsky5_t2v.py |
| kandinsky5 | kandinskylab/Kandinsky-5.0-T2V-Pro-sft-5s-Diffusers |
T2V | basic_kandinsky5_t2v.py |
| kandinsky5 | kandinskylab/Kandinsky-5.0-T2V-Lite-distilled16steps-5s-Diffusers |
T2V | basic_kandinsky5_t2v.py |
| kandinsky5 | kandinskylab/Kandinsky-5.0-T2V-Pro-distilled-5s-Diffusers |
T2V | basic_kandinsky5_t2v.py |
| kandinsky5 | kandinskylab/Kandinsky-5.0-I2V-Lite-5s-Diffusers |
I2V | basic_kandinsky5_i2v.py |
| kandinsky5 | kandinskylab/Kandinsky-5.0-I2V-Pro-sft-5s-Diffusers |
I2V | basic_kandinsky5_i2v.py |
| kandinsky5 | kandinskylab/Kandinsky-5.0-I2V-Pro-distilled-5s-Diffusers |
I2V | basic_kandinsky5_i2v.py |
| kandinsky6 | kandinskylab/Kandinsky-6.0-Pro-5s-Diffuserskandinskylab/Kandinsky-6.0-Pro-sft-5s-Diffusers |
T2V, I2V | basic_kandinsky6_ti2va.py |
| kandinsky6 | kandinskylab/Kandinsky-6.0-Pro-distill-5s-Diffusers |
T2V, I2V | basic_kandinsky6_ti2va.py |
| kandinsky6_sr | kandinskylab/Kandinsky-6.0-VSR-5s-Diffuserskandinskylab/Kandinsky-6.0-VSR-distilled2steps-5s-Diffusers |
— | basic_kandinsky6_sr.py |
| lingbot_video | FastVideo/LingBot-Video-MoE-30B-A3B-Diffusers |
T2V | basic_lingbot_video.py |
| lingbot_video | FastVideo/LingBot-Video-Dense-1.3B-Diffusers |
T2V | basic_lingbot_video.py |
| lingbotworld | FastVideo/LingBot-World-Base-Cam-Diffusers |
I2V | basic_lingbotworld_base_cam.py |
| lingbotworld2 | robbyant/lingbot-world-v2-14b-causal-fast |
I2V | basic_lingbotworld2_causal_fast.py |
| ltx2 | FastVideo/LTX2-Distilled-DiffusersFastVideo/LTX2.3-Distilled-DiffusersFastVideo/LTX-2.3-Distilled-Diffusers |
T2V | basic_ltx2_distilled.py |
| ltx2 | Lightricks/LTX-2.3FastVideo/LTX2.3-baseFastVideo/LTX2.3-Diffusers |
T2V | basic_ltx2.py |
| ltx2 | Lightricks/LTX-2FastVideo/LTX2-baseFastVideo/LTX2-Diffusers |
T2V | basic_ltx2.py |
| mmaudio | FastVideo/MMAudio-large-44k-v2-Diffusers |
V2A, T2A | basic_mmaudio.py |
| matrixgame | FastVideo/Matrix-Game-2.0-Base-Distilled-DiffusersFastVideo/Matrix-Game-2.0-GTA-Distilled-DiffusersFastVideo/Matrix-Game-2.0-TempleRun-Distilled-DiffusersFastVideo/Matrix-Game-2.0-Base-DiffusersFastVideo/Matrix-Game-2.0-GTA-DiffusersFastVideo/Matrix-Game-2.0-TempleRun-Diffusersmignonjia/mg_longtuning_distilled_zeldamignonjia/mg_sf_distilled_zelda_1k_stepsmignonjia/mg_sf_distilled_zeldamignonjia/mg_causal_zeldamignonjia/mg_bidirectional_zelda |
I2V | basic_matrixgame2.py |
| matrixgame | FastVideo/Matrix-Game-3.0-Base-Distilled-Diffusers |
I2V | basic_matrixgame3.py |
| minimax_h3 | MiniMaxAI/MiniMax-H3 |
T2V, I2V | T2VA FL2VA Ref2VA |
| sd35 | stabilityai/stable-diffusion-3.5-medium |
T2I | basic_sd35_t2i.py |
| stable_audio | FastVideo/stable-audio-open-1.0-Diffusers |
T2V | basic_stable_audio.py |
| stable_audio | FastVideo/stable-audio-open-small-Diffusers |
T2V | basic_stable_audio_small.py |
| wan | Wan-AI/Wan2.1-T2V-1.3B-Diffusers |
T2V | basic.py |
| wan | Wan-AI/Wan2.1-T2V-14B-DiffusersFastVideo/Wan2.1-VSA-T2V-14B-720P-Diffusers |
T2V | — |
| wan | Wan-AI/Wan2.1-I2V-14B-480P-Diffusers |
I2V | — |
| wan | Wan-AI/Wan2.1-I2V-14B-720P-Diffusers |
I2V | — |
| wan | weizhou03/Wan2.1-Fun-1.3B-InP-Diffusers |
I2V | — |
| wan | IRMChen/Wan2.1-Fun-1.3B-Control-Diffusers |
— | basic_wan2_2_Fun.py |
| wan | FastVideo/FastWan2.1-T2V-1.3B-DiffusersFastVideo/FastWan2.1-T2V-14B-480P-Diffusers |
T2V | basic_dmd.py |
| wan | Wan-AI/Wan2.2-TI2V-5B-Diffusers |
T2V, I2V | basic_wan2_2_ti2v.py |
| wan | FastVideo/FastWan2.2-TI2V-5B-FullAttn-DiffusersFastVideo/FastWan2.2-TI2V-5B-Diffusers |
T2V, I2V | basic_dmd.py |
| wan | decart-ai/Lucy-Edit-Devdecart-ai/Lucy-Edit-1.1-Dev |
— | basic_lucy_edit.py |
| wan | Wan-AI/Wan2.2-T2V-A14B-Diffusers |
T2V | basic_wan2_2.py |
| wan | Wan-AI/Wan2.2-I2V-A14B-Diffusers |
I2V | basic_wan2_2_i2v.py |
| wan | wlsaidhi/SFWan2.1-T2V-1.3B-Diffusers |
T2V | basic_self_forcing_causal.py |
| wan | rand0nmr/SFWan2.2-T2V-A14B-Diffusers |
T2V | basic_self_forcing_causal_wan2_2_t2v.py |
| wan | FastVideo/SFWan2.2-I2V-A14B-Preview-Diffusers |
I2V | basic_self_forcing_causal_wan2_2_i2v.py |
Note (stable_audio): the Stable Audio Open pipelines generate audio
(StableAudioT2AConfig / StableAudioOpenSmallConfig); they are registered
under the generic T2V workload option in the registry.
Note (MMAudio): the registered Hugging Face model ID is reserved but not
yet public. Follow the MMAudio inference guide
to convert the official weights locally and set MMAUDIO_MODEL_PATH.
Note (MiniMax H3): T2VA, FL2VA, and Ref2VA all generate video with stereo audio. Use the Ref2VA example when passing ordered image, video, or audio references. Distilled Ref2VA PDD students (eight transformer forwards, with reference videos as sparse VSA regions) run through basic_fasth3_omniref_pdd.py; see FastH3 distilled checkpoint schedules.
Note (Kandinsky 6): the two kandinsky6 IDs generate video with audio from
text, optionally plus an image (see the T2IVA guide); the
kandinsky6_sr IDs upscale an existing video (see the
Video SR guide).
Note (Wan-VACE): not currently supported — no VACE pipeline or registered
model ID exists on main
(#1435). The closest
supported path is the Wan2.1-Fun control pipeline
(IRMChen/Wan2.1-Fun-1.3B-Control-Diffusers).
The symbols used have the following meanings:
- ✅ = Full compatibility
- ❌ = No compatibility
- ⭕ = Does not apply to this model
Models x Optimization
The HuggingFace Model ID can be passed directly to
from_pretrained(). FastVideo then uses model-specific default settings for
pipeline initialization and sampling.
Registered models absent from this table have not been validated against these optimizations: absence means untested, not incompatible.
<style> /* Target tables in this section */ #models-x-optimization + p + table { display: block; overflow-x: auto; width: 100%; font-size: 0.85rem; } #models-x-optimization + p + table td, #models-x-optimization + p + table th { text-align: center; white-space: nowrap; padding: 0.5em; } /* First two columns can wrap */ #models-x-optimization + p + table td:nth-child(1), #models-x-optimization + p + table td:nth-child(2) { white-space: normal; min-width: 120px; } #models-x-optimization + p + table td:nth-child(2) code { font-size: 0.75rem; } </style>| Model Name | HuggingFace Model ID | Resolutions | TeaCache | Sliding Tile Attn (Legacy Branch) | Sage Attn | VSA | BSA |
|---|---|---|---|---|---|---|---|
| FastWan2.1 T2V 1.3B | FastVideo/FastWan2.1-T2V-1.3B-Diffusers |
480P | ⭕ | ⭕ | ⭕ | ✅ | ⭕ |
| FastWan2.2 TI2V 5B Full Attn | FastVideo/FastWan2.2-TI2V-5B-FullAttn-Diffusers |
720P | ⭕ | ⭕ | ⭕ | ✅ | ⭕ |
| Wan2.2 TI2V 5B | Wan-AI/Wan2.2-TI2V-5B-Diffusers |
720P | ⭕ | ⭕ | ✅ | ⭕ | ⭕ |
| Lucy Edit Dev 5B*** | decart-ai/Lucy-Edit-Dev |
480P | ⭕ | ⭕ | ⭕ | ⭕ | ⭕ |
| Wan2.2 T2V A14B | Wan-AI/Wan2.2-T2V-A14B-Diffusers |
480P 720P |
❌ | ❌ | ✅ | ⭕ | ⭕ |
| Wan2.2 I2V A14B | Wan-AI/Wan2.2-I2V-A14B-Diffusers |
480P 720P |
❌ | ❌ | ✅ | ⭕ | ⭕ |
| HunyuanVideo | hunyuanvideo-community/HunyuanVideo |
720px1280p 544px960p |
❌ | ✅ | ✅ | ⭕ | ⭕ |
| FastHunyuan | FastVideo/FastHunyuan-diffusers |
720px1280p 544px960p |
❌ | ✅ | ✅ | ⭕ | ⭕ |
| Wan2.1 T2V 1.3B | Wan-AI/Wan2.1-T2V-1.3B-Diffusers |
480P | ✅ | ✅ | ✅ | ⭕ | ⭕ |
| Wan2.1 T2V 14B | Wan-AI/Wan2.1-T2V-14B-Diffusers |
480P, 720P | ✅ | ✅ | ✅ | ⭕ | ⭕ |
| Wan2.1 I2V 480P | Wan-AI/Wan2.1-I2V-14B-480P-Diffusers |
480P | ✅ | ✅ | ✅ | ⭕ | ⭕ |
| Wan2.1 I2V 720P | Wan-AI/Wan2.1-I2V-14B-720P-Diffusers |
720P | ✅ | ✅ | ✅ | ⭕ | ⭕ |
| Matrix Game 2.0 Base Distilled | FastVideo/Matrix-Game-2.0-Base-Distilled-Diffusers |
352x640 | ⭕ | ⭕ | ⭕ | ⭕ | ⭕ |
| Matrix Game 2.0 GTA Distilled | FastVideo/Matrix-Game-2.0-GTA-Distilled-Diffusers |
352x640 | ⭕ | ⭕ | ⭕ | ⭕ | ⭕ |
| Matrix Game 2.0 TempleRun Distilled | FastVideo/Matrix-Game-2.0-TempleRun-Distilled-Diffusers |
352x640 | ⭕ | ⭕ | ⭕ | ⭕ | ⭕ |
| Matrix Game 3.0 Base Distilled | FastVideo/Matrix-Game-3.0-Base-Distilled-Diffusers |
720x1280 | ⭕ | ⭕ | ⭕ | ⭕ | ⭕ |
| GEN3C Cosmos 7B | FastVideo/GEN3C-Cosmos-7B-Diffusers |
704px1280p | ❌ | ❌ | ❌ | ⭕ | ⭕ |
Apple Silicon native runtime
| Release path | Model | Mode | Validated hardware | Status |
|---|---|---|---|---|
| MLX FastMetal T2V 1.3B | FastVideo/FastMetal-1.3B-QAD |
480x832, 81 frames, 3-step DMD, INT8 DiT + TAEHV decode | Apple M4 Max, 16 GB+ unified memory | Released |
| MLX FastMetal T2V 5B | FastVideo/FastMetal-5B-QAD |
480p / 720p, 81 frames, 3-step DMD, INT8 DiT + TAEHV decode; optional --fast, --fast-spatial, --refine. The checked-in example is T2V. CUDA Wan2.2 TI2V 5B is the image-capable path. |
Apple M4 Max, 16 GB+ unified memory | Released |
| MLX FastMetal T2V 14B | FastVideo/FastMetal-14B-QAD |
480p / 720p, 81 frames, 3-step DMD, INT8 DiT + TAEHV decode | Apple M4 Max, 36 GB+ unified memory | Released |
| MLX FastH3 V1 T2VA | FastVideo/FastVideo-Minimax-FastH3-Preview-v0.2 + locally converted DiT |
480p / 720p, 124 frames, 4-step DMD2, INT8/INT6/INT4 weight-only DiT, native video + audio VAE; optional temporal RIFE fast mode; optional spatial fast mode; optional VSA (tile 64/256, exempt/compete) on --include-vsa checkpoints |
Apple M4 Max, 36 GB unified memory | Source runtime; T2VA only |
| MLX FastH3 V2 T2VA | FastVideo/FastVideo-FastH3-8-Step-V2 + locally converted DiT |
480p / 720p, 124 frames, 8-step DMD2, INT8 weight-only DiT with --include-vsa, trained VSA 0.8 / tile 64, native video + audio VAE; AdaLN cache from fastvideo_inference.json |
Apple M4 Max, 36 GB unified memory | Source runtime; T2VA only |
Apple Silicon uses the native MLX runtime. FastMetal-QAD is the packaged Wan
release. FastH3 V1 and FastH3 V2 use a source checkout and local DiT conversion.
CUDA FastWan-QAD (FastVideo/FastWan-QAD-1.3B,
FastVideo/FastWan-QAD-FP8-1.3B) is the NVIDIA release. See the
MLX install guide and the
FastMetal-QAD blog.
Note: Wan2.2 TI2V 5B has some quality issues when performing I2V generation. We are working on fixing this issue.
***Lucy Edit Dev uses a non-commercial model license. FastVideo support is focused on inference integration for video editing workflows.
Sliding Tile Attn (Legacy Branch) entries refer to the archived
sta_do_not_delete branch workflow, not active main inference wiring.
Canonical Supported IDs
The authoritative source for model-ID recognition is
fastvideo/registry.py. If a model ID is registered there, FastVideo can
resolve default pipeline and sampling configuration for it.
Note (GEN3C): The official nvidia/GEN3C-Cosmos-7B repo provides a raw
model.pt checkpoint. Use a Diffusers-format repo (for example,
FastVideo/GEN3C-Cosmos-7B-Diffusers) or convert locally with
scripts/checkpoint_conversion/convert_gen3c_to_fastvideo.py.
Hardware and OS
Per the installation guides:
- NVIDIA GPU (x86_64) — CUDA 12.6 or 13.0; see the GPU install guide.
- NVIDIA DGX Spark (GB10, aarch64) — CUDA 13, from-source kernel build; see the DGX Spark install guide. Two Sparks over QSFP use Ray sequence parallel; see Pair two NVIDIA DGX Sparks.
- Apple silicon — macOS 14 or newer; MLX runtime for FastMetal-QAD, FastH3
V1, and FastH3 V2. See the
MLX install guide. The older
basic_mps.pydemo is PyTorch MPS only.
Optimization-specific hardware constraints (e.g. STA requiring Hopper) are listed under Special requirements.
Special requirements
Sliding Tile Attention
- Full STA pipeline usage is on the archived branch: https://github.com/hao-ai-lab/FastVideo/tree/sta_do_not_delete
- STA currently requires Hopper GPUs (H100s).
Matrix Game 2.0
- Image-to-video game world models with keyboard/mouse control input
- Three variants available: Base (universal), GTA, and TempleRun
- Each variant has different keyboard dimensions for control inputs