[ltx-pipelines] i2v at 1216x832 crashes DiffVAE decoder with illegal memory access (t2v same res OK, i2v 704/768 OK)

Aperta
  • Qualcosa non va?
#288 2 commenti 0 reazioni 0 assegnatari Vedi su GitHub

Nessuno ha ancora preso questa issue.

Valutazione

Difficoltà
4/5
Tempo stimato
3-5 giorni
Idoneità per principianti
18/100
Tipo di issue
Bug
Chiarezza
Abbastanza chiara
Stato di attività
Tranquilla
Stack tecnologico
python, pytorch
Ambito
ai, machine-learning

Direzione di ricerca

The crash sits in the chunked decoder path under ltx_core/model/video_vae/transformer/chunked/ — start by reading context.py (_upsample_then_ctx, deferred) and block.py:forward_x_ctx to see how tiles and context are shaped at 1216x832. Reproduce with the given command on a free GPU, or with a reduced harness that runs only the VAE decode on a latent of that shape, and instrument the tile boundaries at that specific resolution. Done means i2v at 1216x832 completes without the cuBLAS/illegal-address failure, or the pipeline rejects the shape with a clear error.

Scritto dal modello di indicizzazione a partire dal testo della issue.

Descrizione

Summary

ltx_pipelines.distilled with image conditioning at 1216x832 deterministically crashes in the DiffVAE decoder with cudaErrorIllegalAddress. Same resolution without --image works, and the same conditioning at 704/768 works — it is only the combination.

832 is a legal size (64x13) and passes the pipeline's own divisible-by-64 check.

Repro

python -m ltx_pipelines.distilled \
  --transformer-path ltx-2.5-22b-distilled-transformer-nvfp4.safetensors \
  --quantization nvfp4-prequant \
  --text-encoder-path gemma4-12b-with-proj-ltx-2.5-bf16.safetensors \
  --video-vae-path ltx-2.5-video-vae-bf16.safetensors \
  --audio-vae-path ltx-2.5-audio-vae-bf16.safetensors \
  --duration-head-path ltx-2.5-duration-head-bf16.safetensors \
  --spatial-upsampler-path ltx-2.3-spatial-upscaler-x2-1.1.safetensors \
  --image cond_1216x832.png 0 1.0 \
  --width 1216 --height 832 --num-frames 97 --seed 101 \
  --prompt "The camera slowly pushes in over the hillside city" \
  --output-path out.mp4

Isolation

config result
t2v 1216x832 (no --image) OK
i2v 1216x704 OK
i2v 1216x768 OK
i2v 1216x832 FAIL 3/3 (seeds 101/102/103)
i2v 1216x736 / 1216x800 graceful ValueError (not divisible by 64) — expected

Traceback (tail)

ltx_core/model/video_vae/transformer/chunked/block.py:51  in forward_x_ctx
ltx_core/model/video_vae/transformer/chunked/context.py:77 in deferred
ltx_core/model/video_vae/transformer/chunked/context.py:27 in _upsample_then_ctx
RuntimeError: CUDA error: CUBLAS_STATUS_INTERNAL_ERROR when calling cublasGemmStridedBatchedEx(...)
torch.AcceleratorError: CUDA error: an illegal memory access was encountered

Failure is in the decoder's chunked upsample path, not the sampler — looks like a tile-boundary issue at this specific latent shape. Possibly related to #277 (DiffVAE tiling at a boundary), different symptom.

Environment

  • LTX-2 @ 400fd31, ltx-pipelines/ltx-core/ltx-kernels 1.2.0
  • torch 2.13.0+cu132, natten 0.21.7+torch2130cu132 (pinned wheel), CUDA 13.2
  • NVIDIA RTX PRO 6000 Blackwell (sm120), driver 595.45.04
  • Distilled NVFP4 transformer, --quantization nvfp4-prequant
  • GPU otherwise idle (ruled out VRAM starvation — a separate tenant on the card produced OOM/cuBLAS errors that looked similar; these runs were on a free card)

Aside

Model card says width/height must be divisible by 32; the two-stage pipeline enforces 64. Worth aligning the docs.

Workaround: keep i2v at 1216x704 or 1216x768.

Lingua principale
Python
Stelle
10k
Fork
1.6k
Nessuna PR unita negli ultimi 30g

Preparare l'ambiente

Questo progetto non fornisce container di sviluppo, Dockerfile né guida per i contributori, quindi l'ambiente è a tuo carico: parti dal suo README e consulta la nostra guida al primo contributo per i passaggi generali.

Come iniziare

  1. Leggi tutta la issue e poi la guida ai contributi del progetto.
  2. Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
  3. Fai un fork del repository e lavora su un branch.
  4. Apri una pull request che faccia riferimento al numero della issue.

Altre issue di Lightricks/LTX-2

Tutte le issue di Lightricks/LTX-2

Issue simili

Altre issue su Python

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.