ltx2.3-10eros-nvfp4

LTX2.3-10Eros v1.5 (NVFP4)

NVFP4 quantized version of TenStrip/LTX2.3-10Eros (v1.5), a fine-tune/layer-scaled merge built on SulphurAI/Sulphur-2-base, itself a variant of Lightricks/LTX-2.3, for image-to-video generation.

[!NOTE] Not-for-all-audiences: this model is tuned for explicit motion/anatomy generation, per the upstream model card.

Files

Only the v1.5 checkpoint is converted here (repo name kept as plain "eros" since that's the only version quantized so far — other upstream versions, e.g. 1.2/1.3/1.4, are not included).

Requirements

NVFP4 (FP4 E2M1 block quantization) requires a Blackwell GPU (SM >= 10.0/12.0) for inference (e.g. RTX 50-series, B100/B200). Converted and tested on an RTX 5070 Ti.

Load with ComfyUI (or any loader supporting .comfy_quant metadata for NVFP4 weights) as a checkpoint — bundles the transformer, video VAE, and audio VAE, same layout as the upstream bf16 checkpoint.

Per the upstream model card, v1.5 pairs best with a "hybrid" DMD/distill LoRA to make full use of its motion variety (e.g. TenStrip/LTX2.3_DMD_Lora's LTX2.3_DMD_hybrid_v2.safetensors, or maximsobolev275/LTX2.3-DMD-Distill-Lora-Hybrids).

Quantization method

Quantized with the convert-to-quant (ctq) CLI tool using NVFP4 (FP4 E2M1) block quantization, Comfy-quant format, and the built-in --ltxv2 exclusion filter (keeps select transformer blocks high-precision and excludes the VAE and vocoder from quantization — matching LTX-2's mixed-modal audio/video architecture). Same method used for our Sulphur-2-base NVFP4 conversion.

Steps

1. Download the checkpoint (single file, no sharding):

   hf download TenStrip/LTX2.3-10Eros --include "10Eros_v1.5_bf16.safetensors"
   

2. Quantize directly (single file, no shard merge needed):

   uv run --with convert-to-quant --with triton --with safetensors --with tqdm --with numpy ctq \
     -i 10Eros_v1.5_bf16.safetensors \
     -o 10Eros_v1.5_nvfp4.safetensors \
     --ltxv2 --nvfp4 --comfy_quant --save-quant-metadata --simple --low-memory
   

3. Delete the raw bf16 file, keeping only the quantized output.

10Eros_v1.5_nvfp4.safetensors: 9643 tensors, 1232 layers quantized (verified via the _quantization_metadata header) — same exclusion profile/tensor counts as the Sulphur-2-base conversion, since both share the same LTX-2.3 architecture.

Flags used:

• --nvfp4 — FP4 E2M1 block quantization
• --comfy_quant — Comfy quantization tensor/metadata format
• --ltxv2 — LTXv2 exclusion filter (keep select transformer blocks high-precision, exclude vae/vocoder)
• --simple — skip SVD optimization, use simple quantization
• --save-quant-metadata — embed _quantization_metadata in the safetensors header
• --low-memory — stream tensors during quantization to reduce RAM usage

License

LTX-2 Community License Agreement (included in this repo), inherited from SulphurAI/Sulphur-2-base / Lightricks/LTX-2.3.

Additionally, per the creator's (TenStrip) terms for the base 10Eros model:

By downloading, accessing, or using the 10Eros Model (the "Model"), you ("User") agree to the following terms:

1. Permitted Use Only — The Model may only be used with inputs of fictional or non-real characters. Any use involving real persons, celebrities, public figures, or identifiable individuals is strictly prohibited ("Out-of-Scope Use").
2. Disclaimer of Liability — TenStrip (the creator of the Model), together with its affiliates (CivitAI.com / CivitAI.red, Lightricks Ltd., and Sulphur Team), collectively and individually disclaim all liability for any direct, indirect, incidental, consequential, special, or punitive damages arising from your use or misuse of the Model. This includes, without limitation, any claims related to copyright infringement, right of publicity violations, defamation, privacy violations, deepfakes, illegal activities, or any harm caused by outputs generated by the Model.
3. Out-of-Scope Uses Forbidden — All Out-of-Scope Uses are explicitly forbidden. The creator and affiliates bear no responsibility for any outputs generated from prohibited inputs. Your actions with this generative Model are solely your own.
4. Assumption of Risk — You assume all risks associated with the use of the Model. Any use outside the permitted fictional/non-real character inputs further releases TenStrip and all affiliates from any and all liability.
5. Indemnification — You agree to indemnify, defend, and hold harmless TenStrip and its affiliates from any and all claims, losses, liabilities, damages, and expenses (including reasonable attorneys' fees) arising from your use or misuse of the Model.
6. No Warranties — The Model is provided "AS IS" with no warranties of any kind, express or implied.

This agreement is governed by the laws of the United States. If you do not agree to these terms, do not download or use the Model.