ltx-community
/

LTX-2.3-Transformer-GroupC-STG-sm120-cu130-rb3

@@ -1,161 +1,53 @@
 ---
 tags:
 - ahead-of-time
 - pytorch
 library_name: diffusers
 ---
-> [!NOTE]
-> This **README** has been auto-generated by the **HF Job** run linked below
-> and the whole repository is a reproducible artifact of this Job
-# Ahead-of-time repository
-AoT repos contain **pre-compiled binaries** of PyTorch models, enabling:
-- fast startup times (no `torch.compile` needed)
-- significant **speedup**
-- **ZeroGPU** compatibility
-## How to use
-``` python
-import os
-import tempfile
-import numpy as np
-import torch
-import spaces
-from PIL import Image
 from diffusers import LTX2InContextPipeline
-from diffusers.pipelines.ltx2.pipeline_ltx2_ic_lora import LTX2ReferenceCondition
-from diffusers.pipelines.ltx2.utils import DISTILLED_SIGMA_VALUES
-# base == distilled in architecture, so one compiled graph serves both; distilled is
-# what most demos use. The AOTI package is weight-agnostic, so this base graph also
-# serves any FUSED LoRA (fuse_lora before aoti_load on the Space).
-MODEL_ID = os.environ.get("LTX_MODEL_ID", "diffusers/LTX-2.3-Distilled-Diffusers")
-pipe = LTX2InContextPipeline.from_pretrained(MODEL_ID, torch_dtype=torch.bfloat16)
-pipe.to("cuda")
-pipe.vae.enable_tiling()
-spaces.aoti_load(
-    module=pipe.transformer,
-    repo_id='linoyts/LTX-2.3-Transformer-GroupC-STG-sm120-cu130-rb3',
-)
-```
-## How to reproduce or customize
-``` bash
-hf jobs uv run job.py --flavor rtx-pro-6000 --image pytorch/pytorch:2.9.1-cuda13.0-cudnn9-devel --secrets HF_TOKEN
-```
-## Samples
-| Before compilation (0.42s) | After compilation (0.33s) |
-|---|---|
-| <video src="https://huggingface.co/linoyts/LTX-2.3-Transformer-GroupC-STG-sm120-cu130-rb3/resolve/main/samples/before/video.mp4" controls></video> | <video src="https://huggingface.co/linoyts/LTX-2.3-Transformer-GroupC-STG-sm120-cu130-rb3/resolve/main/samples/after/video.mp4" controls></video> |
-Speedup: **1.27x**
-## Environment
-<details><summary>Click to expand</summary>
-```
-PyTorch version: 2.12.0+cu130
-Is debug build: False
-CUDA used to build PyTorch: 13.0
-ROCM used to build PyTorch: N/A
-OS: Ubuntu 22.04.5 LTS (x86_64)
-GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
-Clang version: Could not collect
-CMake version: version 4.1.2
-Libc version: glibc-2.35
-Python version: 3.10.19 (main, Oct 31 2025, 23:02:46) [Clang 21.1.4 ] (64-bit runtime)
-Python platform: Linux-6.12.88-119.157.amzn2023.x86_64-x86_64-with-glibc2.35
-Is CUDA available: True
-CUDA runtime version: 13.0.48
-CUDA_MODULE_LOADING set to:
-GPU models and configuration: GPU 0: NVIDIA RTX PRO 6000 Blackwell Server Edition
-Nvidia driver version: 580.159.03
-cuDNN version: Could not collect
-Is XPU available: False
-HIP runtime version: N/A
-MIOpen runtime version: N/A
-Is XNNPACK available: True
-Caching allocator config: N/A
-CPU:
-Architecture:                            x86_64
-CPU op-mode(s):                          32-bit, 64-bit
-Address sizes:                           46 bits physical, 48 bits virtual
-Byte Order:                              Little Endian
-CPU(s):                                  192
-On-line CPU(s) list:                     0-191
-Vendor ID:                               GenuineIntel
-Model name:                              Intel(R) Xeon(R) Platinum 8559C
-CPU family:                              6
-Model:                                   207
-Thread(s) per core:                      2
-Core(s) per socket:                      48
-Socket(s):                               2
-Stepping:                                2
-BogoMIPS:                                4800.00
-Flags:                                   fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq monitor ssse3 fma cx16 pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch cpuid_fault ssbd ibrs ibpb stibp ibrs_enhanced fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves avx_vnni avx512_bf16 wbnoinvd ida arat avx512vbmi umip pku ospke waitpkg avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq rdpid cldemote movdiri movdir64b md_clear serialize amx_bf16 avx512_fp16 amx_tile amx_int8 flush_l1d arch_capabilities
-Hypervisor vendor:                       KVM
-Virtualization type:                     full
-L1d cache:                               4.5 MiB (96 instances)
-L1i cache:                               3 MiB (96 instances)
-L2 cache:                                192 MiB (96 instances)
-L3 cache:                                640 MiB (2 instances)
-NUMA node(s):                            2
-NUMA node0 CPU(s):                       0-47,96-143
-NUMA node1 CPU(s):                       48-95,144-191
-Vulnerability Gather data sampling:      Not affected
-Vulnerability Indirect target selection: Not affected
-Vulnerability Itlb multihit:             Not affected
-Vulnerability L1tf:                      Not affected
-Vulnerability Mds:                       Not affected
-Vulnerability Meltdown:                  Not affected
-Vulnerability Mmio stale data:           Not affected
-Vulnerability Reg file data sampling:    Not affected
-Vulnerability Retbleed:                  Not affected
-Vulnerability Spec rstack overflow:      Not affected
-Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
-Vulnerability Spectre v1:                Mitigation; usercopy/swapgs barriers and __user pointer sanitization
-Vulnerability Spectre v2:                Mitigation; Enhanced / Automatic IBRS; IBPB conditional; PBRSB-eIBRS SW sequence; BHI BHI_DIS_S
-Vulnerability Srbds:                     Not affected
-Vulnerability Tsa:                       Not affected
-Vulnerability Tsx async abort:           Not affected
-Vulnerability Vmscape:                   Not affected
-Versions of relevant libraries:
-[pip3] Could not collect
-[conda] numpy                        2.3.4            py311h2e04523_0       conda-forge
-[conda] nvidia-cublas                13.0.0.19        pypi_0                pypi
-[conda] nvidia-cuda-cupti            13.0.48          pypi_0                pypi
-[conda] nvidia-cuda-nvrtc            13.0.48          pypi_0                pypi
-[conda] nvidia-cuda-runtime          13.0.48          pypi_0                pypi
-[conda] nvidia-cudnn-cu13            9.13.0.50        pypi_0                pypi
-[conda] nvidia-cufft                 12.0.0.15        pypi_0                pypi
-[conda] nvidia-curand                10.4.0.35        pypi_0                pypi
-[conda] nvidia-cusolver              12.0.3.29        pypi_0                pypi
-[conda] nvidia-cusparse              12.6.2.49        pypi_0                pypi
-[conda] nvidia-cusparselt-cu13       0.8.0            pypi_0                pypi
-[conda] nvidia-nccl-cu13             2.27.7           pypi_0                pypi
-[conda] nvidia-nvjitlink             13.0.39          pypi_0                pypi
-[conda] nvidia-nvtx                  13.0.39          pypi_0                pypi
-[conda] optree                       0.17.0           pypi_0                pypi
-[conda] torch                        2.9.1+cu130      pypi_0                pypi
-[conda] torchaudio                   2.9.1+cu130      pypi_0                pypi
-[conda] torchelastic                 0.2.2            pypi_0                pypi
-[conda] torchvision                  0.24.1+cu130     pypi_0                pypi
-[conda] triton                       3.5.1            pypi_0                pypi
 ```
-</details>
-## Job run
-- [linoyts/6a3288ccfb114ff24a3885f5](https://huggingface.co/jobs/linoyts/6a3288ccfb114ff24a3885f5)

 ---
 tags:
 - ahead-of-time
 - pytorch
+- ltx-2.3
+- zerogpu
 library_name: diffusers
+base_model: diffusers/LTX-2.3-Diffusers
 ---
+# LTX-2.3 Transformer — AOTI build (in-context, **STG-capable**)
+Ahead-of-time **precompiled** `transformer_blocks` of `LTX2VideoTransformer3DModel` for
+**ZeroGPU (sm120 / cu130)**. Graph only (no weights) → works with base **and** distilled
+LTX-2.3 and any **fused** LoRA. Dynamic over video/audio token counts.
+This build compiles the **perturbation (STG) path as always-on tensor math**
+(`torch.lerp(value, hidden_states, perturbation_mask)`), so ONE graph serves both
+spatio-temporal-guidance (STG) and non-STG: `lerp(·, ·, ones)` is a no-op, and the real
+mask blends at the STG block. Use this for the **base-model demos that keep STG**
+(`stg_scale>0`, default `spatio_temporal_guidance_blocks=[28]`) — e.g. beard-removal,
+day-to-night, reference-sheet. (Non-STG demos can use the plain Group A repo.)
+## Use it (ZeroGPU) — load at the **root module level** + a small STG wrapper
+```python
+import spaces, torch
 from diffusers import LTX2InContextPipeline
+pipe = LTX2InContextPipeline.from_pretrained(
+    "diffusers/LTX-2.3-Diffusers", torch_dtype=torch.bfloat16).to("cuda")
+pipe.load_lora_weights(my_lora_state_dict, adapter_name="x")
+pipe.fuse_lora(lora_scale=1.0); pipe.unload_lora_weights()
+spaces.aoti_load(module=pipe.transformer, repo_id="ltx-community/LTX-2.3-Transformer-GroupC-STG-sm120-cu130-rb3")
+# the compiled graph always runs the perturbation lerp, so feed a no-op ones mask when the
+# transformer passes None (non-STG blocks / main pass); the STG pass still passes the real
+# mask to block 28. Also force all_perturbed=False (the python skip-attention shortcut is gone).
+for _blk in pipe.transformer.transformer_blocks:
+    _c = _blk.forward
+    def _fwd(*a, _c=_c, **kw):
+        if kw.get("perturbation_mask", None) is None:
+            _h = kw["hidden_states"]
+            kw["perturbation_mask"] = torch.ones((_h.shape[0],1,1), device=_h.device, dtype=_h.dtype)
+        kw["all_perturbed"] = False
+        return _c(*a, **kw)
+    _blk.forward = _fwd
+@spaces.GPU
+def generate(*args, **kwargs):
+    return pipe(*args, **kwargs)
 ```
+Public repo (graph only) → no token. Built with the bundled `job.py` (env `LTX_STG=1`).