LTX-2.3-Native-Transformer-GroupA (native LTX-2 backend)

AOTI precompiled graph of BasicAVTransformerBlock (submodule transformer_blocks) for the native ltx_core LTXModel (AudioVideo, distilled IC-LoRA path).

  • Backend: github.com/Lightricks/LTX-2 @ ae855f8538843825f9015a419cf4ba5edaf5eec2
  • Compiled on: NVIDIA RTX PRO 6000 Blackwell Server Edition sm120 / cu130, torch 2.12.0+cu130
  • Capture res 1024x1024, 33f. Correctness vs eager: rel max|Δ| video=1.48e-02

Consumption: aoti_load onto transformer.velocity_model, then patch _process_transformer_blocks to flatten TransformerArgs -> compiled flat block -> replace(x=...). See job.py / LTX2_NATIVE_AOTI_PROGRESS.md.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Collections including developerjeremylive/LTX-2.3-Native-Transformer-GroupA-sm120-cu130-r20-etheroi