wildminder/awesome-ltx2
Public
Star 历史趋势
数据来源: GitHub API · 生成自 Stargazers.cn
README.md

Awesome LTX-2

A curated list of models, text encoders, and tools for the LTX-2 video generation suite.

ltx-logo

Telegram X

Table of Contents

Intro

▓ Apps & Tools

LTX2.3-Multifunctional

LTX2.3-Multifunctional is a desktop-optimized version of LTX that lowers GPU requirements and simplifies usage. It integrates all features including image-to-video, text-to-video, start/end frames, lip-sync, video enhancement, and image generation into a single application.

Key Features:

  • Lower GPU Requirements: Only needs 24GB VRAM (vs 32GB for standard desktop version)
  • All-in-One Interface: No complex ComfyUI workflows or error-prone nodes
  • Features: T2V, I2V, start/end frames, lip-sync, video enhancement, image generation, LoRA support
  • Multi-Frame Insertion: Two modes for generating long videos
  • Easy Setup: No third-party software required, just install LTX desktop

Downloads & Resources:

▓ Models

LTX-2 models are available in various formats including full weights, transformers-only, and GGUF quantizations for efficient inference.

▣ Checkpoints

VerNamePrecisionSizeDownload
2.5devbf1642.02 GB
2.5devint8convrot21.50 GB
2.5distilledbf1642.02 GB
2.5distilledint8convrot21.50 GB
2.5distillednvfp418.72 GB
2.5pt (pre-trained)bf1643.0 GB
2.5distillednvfp420.6 GB
2.5devw4a8_convrot12.52 GB
2.5distilledw4a8_convrot12.52 GB
2.5distilledfp819.6 GB
2.5distillednvfp417.4 GB
2.5devw4a814.4 GB
2.5distilledw4a814.4 GB
2.5distilledfp821.9 GB
2.5devint8convrot21.64 GB
2.5devnvfp413.57 GB
2.5distilledint8convrot21.64 GB
2.5distillednvfp413.57 GB
2.3devbf1646.1 GB
2.3devfp829.1 GB
2.3devfp829.9 GB
2.3devint829.1 GB
2.3devnvfp421.7 GB
2.3devfp829.1 GB
2.3distilledbf1646.1 GB
2.3distilledfp829.5 GB
2.3distilledfp829.9 GB
2.3distilledint8tensormixed29.1 GB
2.3distillednvfp417.6 GB
2.3distilledmxfp8mixed29.7 GB
2.3distilled 1.1bf1646.1 GB
2.3ltx23_srx fp8_e4m3 experimentalfp823.1 GB
2ltx-2-19b devbf1643.3 GB
2ltx-2-19b devfp827.1 GB
2ltx-2-19b devfp420 GB
2ltx-2-19b distilledbf1643.3 GB
2ltx-2-19b distilledfp827.1 GB
2ltx-2-19b distillednvfp420 GB

Quantized to fp8_e5m2 to support older Triton with older Pytorch on 30 series GPUs. For WangGP in Pinokio

VerNamePrecisionSizeDownload
2ltx-2-19b devfp8_e5m227.1 GB

· · · · · · · · · · · · · ·

❖ silveroxides Quantizations (mxfp8)

Note: The mxfp8mixed quantization requires a custom fork of ComfyUI-Kitchen with mxfp8 support. Standard ComfyUI installations may not support this quantization format.

ModelQuantSizeDownload
ltx-2.3-22b-devint8mixedtensorwise29.2 GB
ltx-2.3-22b-distilledint8tensormixed29.1 GB
ltx-2.3-22b-distilledint8mixedtensorwise29.2 GB
ltx-2.3-22b-distilledmxfp8mixed29.7 GB

· · · · · · · · · · · · · ·

❖ Distilled LoRA

VerRankPrecisionSizeDownload
2.5450int85.06 GB
2.5450int8_lean_convrot4.53 GB
2.5450nvfp42.66 GB
2.5450bf168.90 GB
2.5384bf167.61 GB
2.5256bf165.10 GB
2.5256bf165.10 GB
2.5128bf162.58 GB
2.564bf161.32 GB
2.5128bf162.31 GB
2.572bf161.38 GB
2.3384bf167.61 GB ┊
2.3208bf164.97 GB
2.3159bf163.83 GB
2.3111bf162.74 GB ┊
2.3105bf162.59 GB
2384bf167.67 GB
2242bf164.88 GB
2175bf163.58 GB
2175fp81.79 GB

· · · · · · · · · · · · · ·

❖ TenStrip Distilled LoRA Experiments

Experimental distilled LoRAs optimized for finetunes and I2V workflows. These LoRAs avoid the issues of the massive rank 384 official LoRA which can be counterproductive with conditioned inputs and finetunes.

NameRankModeSizeDownload
distilled v1.136—739 MBTenStrip
distilled v1.172condsafe662 MBTenStrip
distilled72—1.4 GBTenStrip
distilled v1.132condsafe363 MBTenStrip
distilled v1.152condsafe464 MBTenStrip
distilled v1.172energy1.6 GBTenStrip
distilled v1.196energy2.2 GBTenStrip

Notes:

  • Lower rank LoRAs (72 and below) can be used at 1.0 strength safely for I2V first pass, with upscale pass at 0.4-0.5 strength
  • _ceil suffix indicates the dynamic ceiling during reranking
  • _condsafe suffix indicates cross-attention and other conditioning layers have been zeroed for better I2V compatibility
  • The official rank 384 LoRA can actively dampen conditioning signals in I2V workflows; cond_safe versions work much better

Download All LoRAs

· · · · · · · · · · · · · ·

❖ Spatial Upscaler

Required for current two-stage pipeline implementations in this repository. Download to COMFYUI_ROOT_FOLDER/models/latent_upscale_models folder.

VerNameSizeDownload
2.5spatial-upscaler x2 1.00.93 GB ┊
2.3spatial-upscaler x2 1.0996 MB
2.3spatial-upscaler x1.5 1.01.09 GB
2spatial-upscaler x2 1.01.05 GB

· · · · · · · · · · · · · ·

❖ Temporal Upscaler

Required for current two-stage pipeline implementations in this repository. Download to COMFYUI_ROOT_FOLDER/models/latent_upscale_models folder.

VerNameSizeDownload
2.5temporal-upscaler x2 1.00.24 GB ┊
2.3temporal-upscaler x2 1.0262 MB
2temporal-upscaler x2 1.0262 MB

══════════════════════════════════

▣ Merges

Custom merged models combining multiple control signals or specialized configurations.

VerNameDescriptionDownload
2.3ltx-2.3-22b-distilled-1.1-fused-union-controlMerged model combining Canny, Depth, and Pose control signals for unified control

══════════════════════════════════

▣ Finetunes

Community finetuned models based on LTX-2.3 with specialized improvements and optimizations. Each finetune family may include a backbone checkpoint, low-VRAM component splits, GGUF quants, and merged or extracted LoRAs. Variant cell links go directly to the resolve/main safetensors/gguf file when a single canonical asset covers the row.

❖ DaSiWa

High-performance LoRA-integrated checkpoint family based on LTX 2.3. Includes distilled (4-step) and non-distilled (20-30 step) variants. Recommended sampler: Euler + Simple/Normal/Linear_Quadratic.

VerBuildNamePrecisionSizeDownload
2.3DistilledTreasurechest V1fp819.58 GBDaSiWa
2.3DistilledSolsticecoin V2fp828.06 GBDaSiWa
2.3DistilledDragonleap V4int4mixedtensorwise17.10 GBDaSiWa
2.3DistilledDragonleap V4int8tensormixed25.73 GBDaSiWa
2.3Non-DistilledGoldenLace V3fp827.16 GBDaSiWa
2.3Non-DistilledGoldenLace V3nvfp420.24 GBDaSiWa
2.3Non-Distilled GGUFGoldenLace V3Q2_K7.92 GBDaSiWa
2.3Non-Distilled GGUFGoldenLace V3Q3_K_M9.87 GBDaSiWa
2.3Non-Distilled GGUFGoldenLace V3Q4_K_M12.41 GBDaSiWa
2.3Non-Distilled GGUFGoldenLace V3Q5_K_M14.81 GBDaSiWa
2.3Non-Distilled GGUFGoldenLace V3Q6_K17.35 GBDaSiWa
2.3Non-Distilled GGUFGoldenLace V3Q8_021.99 GBDaSiWa

DaSiWa extracted LoRA:

BuildNameLoRA RankSizeDownload
DMD v2 audioLTX2.3_DMD_v2_avgrank86_audio160_L80-D2086 (audio 160)2.16 GBDaSiWa

· · · · · · · · · · · · · ·

❖ 10Eros

I2V-optimised merge using layer scaled merges at different steps. Not a straight weight merge — behaves much nicer than standard LoRA loading and respects prompts.

VerBuildNamePrecisionSizeDownload
2.3Full10Eros v1bf1644.0 GBTenStrip
2.3Full10Eros v1fp827.8 GBTenStrip
2.3transformer-only10Eros v1fp828.2 GBTenStrip
2.3Full10Eros v1.2bf1644.0 GBTenStrip
2.3Full10Eros v1.2fp832.7 GBTenStrip
2.3Full10Eros v1.3bf1644.0 GBTenStrip
2.3Full10Eros v1.3fp827.8 GBTenStrip
2.3Full10Eros v1.4bf1644.0 GBTenStrip
2.3Full10Eros v1.4fp827.8 GBTenStrip
2.3Full10Eros v1.4int8tensormixed27.8 GBTenStrip
2.3Full10Eros v1 INT8 ConvRotint823.51 GBbertbobson

◦ 10Eros GGUF — vantagewithai low-VRAM quants

vantagewithai/LTX2.3-10Eros-GGUF

vantagewithai/LTX2.3-10Eros-GGUF — v1

QuantSizeDownload
Q3_K_M10.36 GBvantagewithai
Q3_K_S9.63 GBvantagewithai
Q4_012.09 GBvantagewithai
Q4_112.95 GBvantagewithai
Q4_K_M13.31 GBvantagewithai
Q4_K_S12.29 GBvantagewithai
Q5_014.21 GBvantagewithai
Q5_115.07 GBvantagewithai
Q5_K_M15.03 GBvantagewithai
Q5_K_S14.01 GBvantagewithai
Q6_K16.55 GBvantagewithai
Q8_021.19 GBvantagewithai
vantagewithai/LTX2.3-10Eros-1.2-GGUF

vantagewithai/LTX2.3-10Eros-1.2-GGUF — v1.2 — Note: files are named 10Eros_v1.210Eros_v1.2-…gguf (upstream double-stamp glitch); we link verbatim.

QuantSizeDownload
Q3_K_M10.36 GBvantagewithai
Q3_K_S9.63 GBvantagewithai
Q4_012.09 GBvantagewithai
Q4_112.95 GBvantagewithai
Q4_K_M13.31 GBvantagewithai
Q4_K_S12.29 GBvantagewithai
Q5_014.21 GBvantagewithai
Q5_115.07 GBvantagewithai
Q5_K_M15.03 GBvantagewithai
Q5_K_S14.01 GBvantagewithai
Q6_K16.55 GBvantagewithai
Q8_021.19 GBvantagewithai
vantagewithai/LTX2.3-10Eros-1.3-GGUF

vantagewithai/LTX2.3-10Eros-1.3-GGUF — v1.3

QuantSizeDownload
Q3_K_M10.36 GBvantagewithai
Q3_K_S9.63 GBvantagewithai
Q4_012.09 GBvantagewithai
Q4_112.95 GBvantagewithai
Q4_K_M13.31 GBvantagewithai
Q4_K_S12.29 GBvantagewithai
Q5_014.21 GBvantagewithai
Q5_115.07 GBvantagewithai
Q5_K_M15.03 GBvantagewithai
Q5_K_S14.01 GBvantagewithai
Q6_K16.55 GBvantagewithai
Q8_021.19 GBvantagewithai
vantagewithai/LTX2.3-10Eros-1.4-GGUF

vantagewithai/LTX2.3-10Eros-1.4-GGUF — v1.4 (latest)

QuantSizeDownload
Q3_K_M10.36 GBvantagewithai
Q3_K_S9.63 GBvantagewithai
Q4_012.09 GBvantagewithai
Q4_112.95 GBvantagewithai
Q4_K_M13.31 GBvantagewithai
Q4_K_S12.29 GBvantagewithai
Q5_014.21 GBvantagewithai
Q5_115.07 GBvantagewithai
Q5_K_M15.03 GBvantagewithai
Q5_K_S14.01 GBvantagewithai
Q6_K16.55 GBvantagewithai
Q8_021.19 GBvantagewithai
vantagewithai/LTX2.3-10Eros-1.5-GGUF

vantagewithai/LTX2.3-10Eros-1.5-GGUF — v1.5 (latest)

QuantSizeDownload
Q3_K_M11.13 GBvantagewithai
Q3_K_S10.34 GBvantagewithai
Q4_012.98 GBvantagewithai
Q4_113.90 GBvantagewithai
Q4_K_M14.30 GBvantagewithai
Q4_K_S13.20 GBvantagewithai
Q5_015.26 GBvantagewithai
Q5_116.18 GBvantagewithai
Q5_K_M16.14 GBvantagewithai
Q5_K_S15.04 GBvantagewithai
Q6_K17.77 GBvantagewithai
Q8_022.76 GBvantagewithai

◦ 10Eros Splits

per-version component split

vantagewithai/LTX2.3-10Eros-Split — v1

ComponentPrecisionSizeDownload
Modelbf1641.03 GBvantagewithai
Modelfp824.45 GBvantagewithai
VAE—1.42 GBvantagewithai
Audio VAE—364.86 MBvantagewithai
Text encoder—2.26 GBvantagewithai

vantagewithai/LTX2.3-10Eros-1.2-Split — v1.2

ComponentPrecisionSizeDownload
Modelbf1641.03 GBvantagewithai
Modelfp829.49 GBvantagewithai
VAE—1.42 GBvantagewithai
Audio VAE—364.86 MBvantagewithai
LoRA (bundled)—662.07 MBvantagewithai

vantagewithai/LTX2.3-10Eros-1.3-Split — v1.3

ComponentPrecisionSizeDownload
Modelbf1641.03 GBvantagewithai
Modelfp824.45 GBvantagewithai
VAE—1.42 GBvantagewithai
Audio VAE—364.86 MBvantagewithai
Text encoder—2.26 GBvantagewithai

vantagewithai/LTX2.3-10Eros-1.4-Split — v1.4 (latest)

ComponentPrecisionSizeDownload
Modelbf1641.03 GBvantagewithai
Modelfp824.45 GBvantagewithai
VAE—1.42 GBvantagewithai
Audio VAE—364.86 MBvantagewithai
Text encoder—2.26 GBvantagewithai

◦ 10Eros Splits — AX1Y2JP transformer-only fork (alternative split for ComfyUI)

VariantDownload
v1.4 transformer_only bf16AX1Y2JP
v1.4 transformer_only fp8mixed_learnedAX1Y2JP
v1.4 video VAE bf16AX1Y2JP

◦ 10Eros Extracted LoRA — maximsobolev275 trained LoRAs

VariantDescriptionDownload
rank-768 family (canonical, v1.4)Author: maximsobolev275. Training LoRAs directly extracted from the 10Eros v1.4 merge (rank 768); lower-rank rerolls for v1.2 / v1.3 / v1.4 are also published in the same repo.LTX-10Eros-LoRA-r768

· · · · · · · · · · · · · ·

❖ Sulphur-2-base

Repo: SulphurAI/Sulphur-2-base — uncensored video generation model based on LTX 2.3 with built-in prompt enhancer. Merge base for 10Eros. T2V + I2V native.

◦ Main video model

VerBuildPrecisionSizeDownload
2.3devbf1644.0 GBSulphur
2.3devfp827.8 GBSulphur
2.3distilbf1644.0 GBSulphur
2.3distilfp827.8 GBSulphur
2.3distilnvfp418.6 GBSulphur

◦ vantagewithai component split — vantagewithai/Sulphur-2-Base-Split

ComponentPrecisionSizeDownload
model (sulphur_dev_bf16_model)bf1640.06 GBvantagewithai
model (sulphur_dev_model_fp8mixed)fp823.87 GBvantagewithai
model (sulphur_distil_bf16_model)bf1640.06 GBvantagewithai
vae—1.38 GBvantagewithai
audio_vae—348.0 MBvantagewithai
Abiray GGUF quants

◦ Abiray GGUF quants — Abiray/Sulphur-2-base-GGUF

BuildPrecisionSizeDownload
sulphur_devbf1640.09 GBAbiray
sulphur_devQ3_K_M10.36 GBAbiray
sulphur_devQ3_K_S9.63 GBAbiray
sulphur_devQ4_012.09 GBAbiray
sulphur_devQ4_K_M13.31 GBAbiray
sulphur_devQ4_K_S12.29 GBAbiray
sulphur_devQ5_014.21 GBAbiray
sulphur_devQ5_K_M15.04 GBAbiray
sulphur_devQ5_K_S14.01 GBAbiray
sulphur_devQ6_K16.55 GBAbiray
sulphur_devQ8_021.19 GBAbiray

◦ Sulphur LoRAs

VerBuildSizeDownload
2.3sulphur_lora_rank_7689.79 GBSulphur
2.3 (experimental)sulphur_experimental_lora_v113.87 GBSulphur

◦ Prompt Enhancer

VariantPrecisionSizeDownload
Censoredbf16879.01 MBSulphur
CensoredQ8_09.09 GBSulphur
Uncensoredbf16879.01 MBSulphur
UncensoredQ8_09.33 GBSulphur

· · · · · · · · · · · · · ·

❖ JoyAI-Echo Surgical

Surgical finetune based on jdopensource/JoyAI-Echo by joeygambino. Combined "echoVid + ltxAud" surgical variant — jointly fine-tuned for both video generation and audio on top of LTX-2.3. A LoRA extracted from JoyAI-Echo (TenStrip's LTX2.3_JoyAI_Lora_Extracted) is listed separately under ### ▣ Special.

VerBuildNamePrecisionSizeDownload
2.3unsplit (full DiT)echoVid-ltxAud surgicalbf1642.97 GBjoeygambino
2.3unsplit (full DiT)echoVid-ltxAud surgicalfp823.41 GBjoeygambino
2.3unsplit (full DiT)echoVid-ltxAud surgicalint8tensormixed27.15 GBjoeygambino
2.3transformer-onlyechoVid-ltxAud surgicalint8tensormixed26.07 GBjoeygambino

◦ JoyAI-Echo Surgical — GGUF (low-VRAM quants)

joeygambino/joyai-echo-ltx23-echoVid-ltxAud-surgical-gguf

joeygambino/joyai-echo-ltx23-echoVid-ltxAud-surgical-gguf — Surgical DiT GGUF quants by joeygambino.

VerBuildNamePrecisionSizeDownload
2.3GGUFechoVid-ltxAud surgicalQ5_015.54 GBjoeygambino
2.3GGUFechoVid-ltxAud surgicalQ8_023.13 GBjoeygambino

· · · · · · · · · · · · · ·

❖ PinkCherry NSFW

Repo: SexGod1979/PinkCherry_NSFW_LTX23 — uncensored NSFW LTX-2.3 finetune family (NSFW content only — do not use for clean content). Pairs with the official LTX-2.3 distilled LoRA 384.

VariantPrecisionSizeDownload
v1.3 devbf1646.14 GBPinkCherry
v1.3 devfp827.62 GBPinkCherry
v1.5 devbf1646.14 GBPinkCherry
v1.5 devint827.64 GBPinkCherry
v1.6 devbf1646.14 GBPinkCherry
v1.6 devfp827.62 GBPinkCherry
v1.6 devint827.64 GBPinkCherry
v1.7-alpha devbf1646.14 GBPinkCherry
v1.7-alpha devfp827.62 GBPinkCherry
v1.7-alpha devint827.64 GBPinkCherry
v1.8 devbf1646.14 GBPinkCherry
v1.8 devfp827.62 GBPinkCherry
v1.8 devint827.64 GBPinkCherry

◦ PinkCherry GGUF — low-VRAM quants by SexGod1979

v1.7-alpha / v1.8 GGUF
QuantBuildSizeDownload
Q5_K_Mv1.7-alpha15.93 GBPinkCherry
Q6_Kv1.7-alpha17.77 GBPinkCherry
Q5_K_Mv1.815.93 GBPinkCherry
Q8_0v1.822.76 GBPinkCherry

❖ Elastic

Elastic is a TensorRT-engine distribution of a LoRA-integrated distilled FP8 T2V variant of LTX-2.3, packaged by TheStageAI as .qlip shard files for H100 GPUs.

BuildNamePrecisionSizeDownload
distil + LoRA T2VElastic — H100fp8~19 GB (49 .qlip shards)TheStageAI

· · · · · · · · · · · · · ·

❖ SpatialAV2AV (BingoG)

Publicly-released audio→audio-video training checkpoints from BingoG/LTX-2-SpatialAV2AV-Checkpoints for spatial-audio conditioned video generation. Two curriculum layouts (E4 Core Dual + Dynamic; E6 reserved for Full) at 480p cap, 113 frames at 25 fps.

CurriculumStepSizeDownload
E4 Core (Dual + Dynamic)40036.22 GBBingoG
E4 Core (Dual + Dynamic)100036.22 GBBingoG
E6 Full100036.22 GBBingoG

Note: these are intermediate training checkpoints, not inference-ready models — useful for fine-tuning experiments and reproducibility only.

══════════════════════════════════

▣ GGUF Quantized Models

These models are optimized for lower memory usage. Note that in ComfyUI, these are typically loaded as transformer-only models.

QuantStack

QuantStack LTX-2.3

ModelQuantSizeDownload
ltx-2.3-22bQ2_K12.4 GBdev ┊ distilled ┊ distilled-1.1
ltx-2.3-22bQ3_K_M14.7 GBdev ┊ distilled ┊ distilled-1.1
ltx-2.3-22bQ3_K_S14 GBdev ┊ distilled ┊ distilled-1.1
ltx-2.3-22bQ4_K_M17.8 GBdev ┊ distilled ┊ distilled-1.1
ltx-2.3-22bQ4_K_S16.7 GBdev ┊ distilled ┊ distilled-1.1
ltx-2.3-22bQ5_K_M19.4 GBdev ┊ distilled ┊ distilled-1.1
ltx-2.3-22bQ5_K_S18.5 GBdev ┊ distilled ┊ distilled-1.1
ltx-2.3-22bQ6_K21 GBdev ┊ distilled ┊ distilled-1.1
ltx-2.3-22bQ8_025.5 GBdev ┊ distilled ┊ distilled-1.1

QuantStack LTX-2

ModelQuantSizeDownload
LTX-2-devQ2_K8.03 GB
LTX-2-devQ3_K_M10.3 GB
LTX-2-devQ3_K_S9.57 GB
LTX-2-devQ4_K_M13.4 GB
LTX-2-devQ4_K_S12.3 GB
LTX-2-devQ5_K_M15 GB
LTX-2-devQ5_K_S14.2 GB
LTX-2-devQ6_K16.6 GB
LTX-2-devQ8_021.1 GB
Unsloth

Unsloth LTX-2.3 GGUF

ModelQuantSizeDownload
ltx-2.3-22bBF1642 GBdev ┊ distilled
ltx-2.3-22bF1642 GBdev ┊ distilled
ltx-2.3-22bQ2_K8.28 GBdev ┊ distilled
ltx-2.3-22bQ3_K_M10.8 GBdev ┊ distilled
ltx-2.3-22bQ3_K_S9.95 GBdev ┊ distilled
ltx-2.3-22bQ4_012.7 GBdev ┊ distilled
ltx-2.3-22bQ4_113.8 GBdev ┊ distilled
ltx-2.3-22bQ4_K_M14.3 GBdev ┊ distilled
ltx-2.3-22bQ4_K_S13.1 GBdev ┊ distilled
ltx-2.3-22bQ5_015.3 GBdev ┊ distilled
ltx-2.3-22bQ5_116.3 GBdev ┊ distilled
ltx-2.3-22bQ5_K_M16.1 GBdev ┊ distilled
ltx-2.3-22bQ5_K_S15.2 GBdev ┊ distilled
ltx-2.3-22bQ6_K17.8 GBdev ┊ distilled
ltx-2.3-22bQ8_022.8 GBdev ┊ distilled
ltx-2.3-22bUD-Q2_K9.5 GBdev ┊ distilled
ltx-2.3-22bUD-Q3_K_M13.5 GBdev ┊ distilled
ltx-2.3-22bUD-Q3_K_S11.4 GBdev ┊ distilled
ltx-2.3-22bUD-Q4_K_M16.5 GBdev ┊ distilled
ltx-2.3-22bUD-Q4_K_S14.2 GBdev ┊ distilled
ltx-2.3-22bUD-Q5_K_M18.3 GBdev ┊ distilled
ltx-2.3-22bUD-Q5_K_S16.3 GBdev ┊ distilled

Unsloth LTX-2.3 GGUF - Distilled 1.1

ModelQuantSizeDownload
ltx-2.3-22bBF1642 GBdistilled-1.1
ltx-2.3-22bF1642 GBdistilled-1.1
ltx-2.3-22bQ2_K7.94 GBdistilled-1.1
ltx-2.3-22bQ3_K_M10.6 GBdistilled-1.1
ltx-2.3-22bQ3_K_S9.74 GBdistilled-1.1
ltx-2.3-22bQ4_K_M14.2 GBdistilled-1.1
ltx-2.3-22bQ4_K_S13 GBdistilled-1.1
ltx-2.3-22bQ5_K_M15.9 GBdistilled-1.1
ltx-2.3-22bQ5_K_S15 GBdistilled-1.1
ltx-2.3-22bQ6_K17.8 GBdistilled-1.1
ltx-2.3-22bQ8_022.8 GBdistilled-1.1
ltx-2.3-22bUD-Q2_K10.9 GBdistilled-1.1
ltx-2.3-22bUD-Q3_K_M13.4 GBdistilled-1.1
ltx-2.3-22bUD-Q4_K_M16.4 GBdistilled-1.1
ltx-2.3-22bUD-Q4_K_S14.1 GBdistilled-1.1
ltx-2.3-22bUD-Q5_K_M18.2 GBdistilled-1.1

Unsloth LTX-2 GGUF

ModelQuantSizeDownload
ltx-2-19b-devBF1637.8 GB
ltx-2-19b-devF1637.8 GB
ltx-2-19b-devUD-Q2_K_L10.1 GB
ltx-2-19b-devUD-Q2_K_XL11.6 GB
ltx-2-19b-devQ2_K8.1 GB
ltx-2-19b-devQ3_K_L10.7 GB
ltx-2-19b-devQ3_K_M10.1 GB
ltx-2-19b-devQ3_K_S9.47 GB
ltx-2-19b-devQ4_011.3 GB
ltx-2-19b-devQ4_112.3 GB
ltx-2-19b-devQ4_K_M12.8 GB
ltx-2-19b-devQ4_K_S11.9 GB
ltx-2-19b-devQ5_013.7 GB
ltx-2-19b-devQ5_114.6 GB
ltx-2-19b-devQ5_K_M14.3 GB
ltx-2-19b-devQ5_K_S13.6 GB
ltx-2-19b-devQ6_K16 GB
ltx-2-19b-devQ8_020.4 GB
Vantage

Vantage AI GGUFs

ModelQuantSizeDownload
ltx-2-19b-devQ3_K_M9.96 GB
ltx-2-19b-devQ3_K_S9.28 GB
ltx-2-19b-devQ4_011.6 GB
ltx-2-19b-devQ4_112.4 GB
ltx-2-19b-devQ4_K_M12.8 GB
ltx-2-19b-devQ4_K_S11.8 GB
ltx-2-19b-devQ5_013.6 GB
ltx-2-19b-devQ5_114.5 GB
ltx-2-19b-devQ5_K_M14.4 GB
ltx-2-19b-devQ5_K_S13.5 GB
ltx-2-19b-devQ6_K15.9 GB
ltx-2-19b-devQ8_020.4 GB
ltx-2-19b-distilledQ3_K_M9.96 GB
ltx-2-19b-distilledQ3_K_S9.28 GB
ltx-2-19b-distilledQ4_011.6 GB
ltx-2-19b-distilledQ4_112.4 GB
ltx-2-19b-distilledQ4_K_M12.8 GB
ltx-2-19b-distilledQ4_K_S11.8 GB
ltx-2-19b-distilledQ5_013.6 GB
ltx-2-19b-distilledQ5_114.5 GB
ltx-2-19b-distilledQ5_K_M14.4 GB
ltx-2-19b-distilledQ5_K_S13.5 GB
ltx-2-19b-distilledQ6_K15.9 GB
ltx-2-19b-distilledQ8_020.4 GB
Abiray LTX-2.5 Distilled GGUF

Abiray/LTX-2.5-Distilled-GGUF

ModelQuantSizeDownload
ltx-2.5-22b-distilledQ3_K_M12.92 GB
ltx-2.5-22b-distilledQ3_K_S12.65 GB
ltx-2.5-22b-distilledQ4_K_M15.69 GB
ltx-2.5-22b-distilledQ4_K_S15.33 GB
ltx-2.5-22b-distilledQ5_K_M18.12 GB
ltx-2.5-22b-distilledQ6_K18.62 GB
ltx-2.5-22b-distilledQ8_023.60 GB
Abiray LTX-2.5 GGUF

Abiray/LTX-2.5-GGUF

ModelQuantSizeDownload
ltx-2.5-22bQ3_K_M9.89 GB
ltx-2.5-22bQ3_K_S9.07 GB
ltx-2.5-22bQ4_K_M13.21 GB
ltx-2.5-22bQ4_K_S12.07 GB
ltx-2.5-22bQ5_K_M14.83 GB
ltx-2.5-22bQ5_K_S14.01 GB
ltx-2.5-22bQ6_K16.55 GB
ltx-2.5-22bQ8_021.19 GB
realrebelai LTX-2.5 GGUFs

realrebelai/LTX-2.5_GGUFs

ModelQuantSizeDownload
ltx-2.5-22b-distilledQ2_K8.23 GB
ltx-2.5-22b-distilledQ3_K_M10.73 GB
ltx-2.5-22b-distilledQ4_K_M14.05 GB
ltx-2.5-22b-distilledQ4_K_S12.90 GB
ltx-2.5-22b-distilledQ5_K_M15.66 GB
ltx-2.5-22b-distilledQ6_K17.38 GB
ltx-2.5-22b-distilledQ8_022.01 GB
vantagewithai LTX-2.5 GGUF

vantagewithai/LTX-2.5-GGUF

ModelQuantSizeDownload
ltx-2.5-22b-devQ2_K12.13 GB
ltx-2.5-22b-devQ3_K_M12.92 GB
ltx-2.5-22b-devQ3_K_S12.65 GB
ltx-2.5-22b-devQ4_015.24 GB
ltx-2.5-22b-devQ4_115.53 GB
ltx-2.5-22b-devQ4_K_M15.69 GB
ltx-2.5-22b-devQ4_K_S15.33 GB
ltx-2.5-22b-devQ5_015.98 GB
ltx-2.5-22b-devQ5_116.26 GB
ltx-2.5-22b-devQ5_K_M18.12 GB
ltx-2.5-22b-devQ5_K_S15.89 GB
ltx-2.5-22b-devQ6_K18.62 GB
ltx-2.5-22b-devQ8_023.60 GB
ltx-2.5-22b-distilledQ2_K12.13 GB
ltx-2.5-22b-distilledQ3_K_M12.92 GB
ltx-2.5-22b-distilledQ3_K_S12.65 GB
ltx-2.5-22b-distilledQ4_015.24 GB
ltx-2.5-22b-distilledQ4_115.53 GB
ltx-2.5-22b-distilledQ4_K_M15.69 GB
ltx-2.5-22b-distilledQ4_K_S15.33 GB
ltx-2.5-22b-distilledQ5_015.98 GB
ltx-2.5-22b-distilledQ5_116.26 GB
ltx-2.5-22b-distilledQ5_K_M18.12 GB
ltx-2.5-22b-distilledQ5_K_S15.89 GB
ltx-2.5-22b-distilledQ6_K18.62 GB
ltx-2.5-22b-distilledQ8_023.60 GB

Special Quantization: PolarQuant Q5

LTX-2.3 (22B) — PolarQuant Q5 is a bit-packed quantization method using Hadamard-Rotated Lloyd-Max Quantization. It achieves optimal Gaussian weight quantization via Hadamard rotation, delivering near-lossless quality with significant size reduction.

Specification image
SpecificationValue
Parameters22B
Transformer Blocks48
Hidden Dimension4096
Layers Quantized1,347 (of 5,947 total tensors)

Compression Statistics:

ComponentOriginal SizePQ5 PackedReduction
Transformer (1,347 layers)37 GB4.6 GB-88%
VAE + Skip (4,600 layers)9.1 GB9.1 GBBF16 kept
Upscalers1.3 GB1.3 GBBF16 kept
Total46.2 GB15 GB-68%
image

Quality Metrics:

  • Cosine Similarity: 0.9986 (near-lossless)
  • Download Size: 15 GB
  • Beats torchao INT4 on perplexity (PPL)

Hardware Requirements:

GPUVRAMStatus
A100 (80 GB)80 GBFull speed
A100 (40 GB)40 GBRecommended
RTX 4090 (24 GB)24 GBWith offloading

Key Features:

  • Mixed precision approach: transformer heavily quantized (-88%) while VAE remains BF16
  • 5-bit bit-packed representation (Q5)
  • 50-65% smaller than original with zero quality loss
  • One-command setup with easy generation wrapper
ModelSizeDownload
LTX-2.3-22B-PolarQuant-Q515 GB

Installation: pip install safetensors huggingface_hub scipy ArXiv Reference: 2603.29078

◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆

▓ Text Encoders

LTX-2/2.3 require Gemma-3-12b variants with text projection layers. LTX-2.5 ships with a Gemma-4-12B text encoder — see the ▣ Gemma-4-12b section below.

▣ Comfy-Org Optimized Encoders

Official and optimized versions for ComfyUI.

Model NameSizeDownload
gemma_3_12B_it24.4 GB
gemma_3_12B_it_fpmixed13.7 GB
gemma_3_12B_it_fp8_scaled13.2 GB
gemma_3_12B_it_fp4_mixed9.5 GB
gemma_3_12B_it-int8tensormixed13.2 GB
gemma_3_12B_it-int8mixedblockwise13.6 GB
gemma_3_12B_it-int8mixedtensorwise14.1 GB
gemma_3_12B_it-int8tensormixed13.2 GB
text_projection_fp81.16 GB
  • gemma_3_12B_it_fpmixed: Experimental quant. Should be better than the fp8 scaled
  • gemma_3_12B_it_fp4_mixed: 90% fp4 layers Note: mxfp8mixed quantization requires a custom fork of ComfyUI-Kitchen with mxfp8 support. Standard ComfyUI setups don't.

· · · · · · · · · · · · · ·

▣ Gemma-3-12b Abliterated

Why Choose Abliterated Encoders?

Standard Gemma models often incorporate safety alignment that "sanitizes" or weakens specific concepts within prompt embeddings. Even when the model doesn't explicitly refuse a request, this internal filtering can dilute creative intent. For LTX-2 video generation, using a standard encoder often results in:

  • Reduced Prompt Adherence: Key stylistic or descriptive terms may be ignored or weakened.
  • Visual Softening: Visual intensity and fine details are often "muted" to fit generic safety profiles.
  • Concept Dilution: Complex or niche creative requests are subtly altered, leading to less faithful representations of your vision.

Abliteration bypasses these restrictive alignment layers, allowing the encoder to translate your prompts into embeddings with maximum fidelity. This ensures LTX-2 receives the most accurate and un-filtered instructions possible.

Gemma-3-12b-Abliterated (FusionCow)

Fixed versions of the abliterated Gemma-3-12b-it model by FusionCow, modified specifically for compatibility with LTX-2. The original model

ModelPrecisionSizeDownload
Gemma ablit fixedbf1623.5 GB
Gemma ablit fixedfp813.8 GB
Sikaworld1990 Gemma-3-12b Abliterated

NVFP4 quantization variants by Sikaworld1990 optimized for Blackwell GPUs.

ModelPrecisionSizeDownload
Gemma-3-12b QAT Abliterated FP4NVFP4-HF12.1 GB
Gemma-3-12b QAT Abliterated FP4NVFP4-Pure8.91 GB
Gemma-3-12b HereticX Abliteratedbf1615 GB
Gemma-3-12b High-Fidelity Abliteratedbf1614.1 GB
  • FP4-HF: High-fidelity mixed precision calibration
  • FP4-Pure: Pure FP4 quantization for maximum compression
  • HereticX: Uncensored variant with maximum prompt fidelity
  • High-Fidelity: Optimized for quality with better detail preservation

· · · · · · · · · · · · · ·

▣ Gemma-3-12b IT Heretic

Models by DreamFast. "Heretic" lineage bypasses alignment/restriction layers in the text encoder so LTX-2/2.3 receives the most faithful prompt embeddings. Two upstream versions (v1, v2) plus an AX1Y2JP ultra-uncensored fork and a 3rd-party mradermacher imatrix GGUF re-quant set.

Heretic v1 — DreamFast (bf16 + fp8 + 8 GGUFs)

Safetensors

ModelPrecisionSizeDownload
Gemma_3_12B_it Hereticbf1623.5 GB
Gemma_3_12B_it Hereticfp812.8 GB

GGUF

QuantSizeDownload
F1622 GB
Q8_012 GB
Q6_K9.0 GB
Q5_K_M7.9 GB
Q5_K_S7.7 GB
Q4_K_M6.8 GB
Q4_K_S6.5 GB
Q3_K_M5.6 GB
Heretic v2 — DreamFast (5 safetensors + 8 GGUFs)

Safetensors

ModelPrecisionSizeDownload
gemma-3-12b-it-heretic-v2bf1623.25 GB
gemma-3-12b-it-heretic-v2fp811.63 GB
gemma-3-12b-it-heretic-v2int8tensormixed12.60 GB
gemma-3-12b-it-heretic-v2mxfp8mixed12.93 GB
gemma-3-12b-it-heretic-v2nvfp47.94 GB

GGUF

QuantSizeDownload
F1622.45 GB
Q8_011.93 GB
Q6_K9.21 GB
Q5_K_M8.05 GB
Q5_K_S7.85 GB
Q4_K_M6.96 GB
Q4_K_S6.61 GB
Q3_K_M5.73 GB
AX1Y2JP Ultra-Uncensored Heretic ComfyUI fp8_scaled

Ultra-uncensored fork of the Heretic encoder, fp8-scaled and ComfyUI-ready. Single safetensors by AX1Y2JP.

ModelPrecisionSizeDownload
gemma-3-12b-it-heretic (ultra-uncensored)fp812.99 GB
mradermacher imatrix GGUF re-quant of Heretic v2 (24 quants)

Third-party imatrix (importance-matrix) re-quantization of DreamFast's Heretic v2 by mradermacher. Covers the full i1-IQ* and i1-Q_K family for low-VRAM use. Sizes verified from the repo file list.

QuantSizeDownload
IQ1_M3.02 GB
IQ1_S2.81 GB
IQ2_M4.11 GB
IQ2_S3.83 GB
IQ2_XS3.66 GB
IQ2_XXS3.36 GB
IQ3_M5.39 GB
IQ3_S5.21 GB
IQ3_XS4.96 GB
IQ3_XXS4.56 GB
IQ4_NL6.57 GB
IQ4_XS6.25 GB
Q2_K4.55 GB
Q2_K_S4.24 GB
Q3_K_L6.18 GB
Q3_K_M5.73 GB
Q3_K_S5.21 GB
Q4_06.59 GB
Q4_17.21 GB
Q4_K_M6.96 GB
Q4_K_S6.61 GB
Q5_K_M8.05 GB
Q5_K_S7.85 GB
Q6_K9.21 GB

▣ Gemma-4-12b (LTX-2.5 Text Encoders)

LTX-2.5 replaces the Gemma-3-12B text encoder with a Gemma-4-12B encoder. Community "heretic" / uncensored forks bypass alignment layers for maximum prompt fidelity in downstream video generation.

ModelPrecisionSizeDownload
gemma4-12b-heretic-ltx-2.5bf1624.46 GB
Gemma-4-12B-it-uncensored-hereticbf1626.26 GB
Gemma-4-12B-it-uncensored-hereticint8convrot13.17 GB
gemma4-12b-ltx2.5-int4int8-mixint4int8mix7.52 GB
gemma4-12b-with-proj-ltx-2.5int815.50 GB
gemma4-12b-with-proj-ltx-2.5int8_lean_convrot15.37 GB
gemma4-12b-with-proj-ltx-2.5nvfp411.20 GB

◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆

▓ Separated Components

Separated LTX2 checkpoint by Kijai and Kijai for LTX-2.3. For alternative way to load the models in Comfy.

▣ Diffusion Models (Transformer Only)

VerNamePrecisionSizeDownload
2.3ltx-2.3-22b devbf1642 GB
2.3ltx-2.3-22b devfp823.5 GB
2.3ltx-2.3-22b devmxfp8_block3224.1 GB
2.3ltx-2.3-22b devfp8_input_scaled25 GB
2.3ltx-2.3-22b distilledbf1642 GB
2.3ltx-2.3-22b distilledfp8_input_scaled23.5 GB
2.3ltx-2.3-22b distilled v2fp8_input_scaled v223.2 GB
2.3ltx-2.3-22b distilledfp823.5 GB
2.3ltx-2.3-22b distilled (experimental)mxfp824.1 GB
2.3ltx-2.3-22b distilled 1.1bf1642 GB
2.3ltx-2.3-22b distilled 1.1fp825.2 GB
2.3ltx-2.3-22b distilled 1.1 (experimental)mxfp824.1 GB
2.3ltx-2.3-22b devint8tensormixed20.51 GB
2.3ltx-2.3-22b distilled 1.1int8tensormixed20.51 GB
2.3ltx-2.3-22b distilled v3fp8_input_scaled23.86 GB
2ltx-2-19b devbf1637.8 GB
2ltx-2-19b devfp821.6 GB
2ltx-2-19b devfp414.5 GB
2ltx-2-19b distilledbf1637.8 GB
2ltx-2-19b distilledfp821.6 GB

[!NOTE]
input_scaled additionally have activation scaling, and are set to run with fp8 matmuls on supported hardware (roughly 40xx and later Nvidia GPUs).

▣ VAE (Video & Audio)

VerComponentPrecisionSizeDownload
2.5Video VAEBF161.37 GB
2.5Video VAE (conv)BF161.35 GB
2.5Audio VAEBF160.34 GB
2.3Video VAEBF161.45 GB ┊
2.3Cinematic Video VAEBF161.38 GB
2.3Pruna Video VAEBF161.27 GB
2.3TAE (tiny autoencoder)BF1622 MB
2.3Audio VAEBF16365 MB ┊
2Video VAEBF162.45 GB
2Audio VAEBF16218 MB

▣ Embedding Connectors & Text Projection

VerNamePrecisionSizeDownload
2.3Embeddings Connectors devbf162.31 GB ┊
2.3Embeddings Connectors distilledbf162.31 GB
2Connector devbf162.86 GB
2Connector distilledbf162.86 GB

◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆

▓ LoRA

▣ Enchancer, special

  • Lightricks LTX-2.3

    • LipDub IC-LoRA - Enables lip dubbing on top of LTX-2.3 for video dubbing via joint audio-visual diffusion (based on JustDubIt research)

  • OmerHagawa

  • systms

    • SYSTMS FLW IC-LoRA - Seamless shot-to-shot transitions IC-LoRA with trigger word FLW, uses gray frames (RGB 127,127,127) between clips

  • LTX-2.3-IC-LoRA-Colorizer by DoctorDiffusion (331 MB) - Colorize black and white videos

  • JUST-DUB-IT

  • Best-Face-Swap-Video

  • Image-to-Video Adapter LoRA

    • Original by MachineDelusions
    • siraxe variant - Stripped audio layers + rank64 compressed (2.62 GB, 655 MB rank64 bf16)

  • Lightricks LTX-2.3

    • HDR - Enables 16-bit HDR video generation and converts SDR video to HDR using LogC3 transform for extended dynamic range
    • Union Control - Unified IC-LoRA combining Canny + Depth + Pose control signals for multi-signal video generation conditioning
    • Motion Track Control - Guides object motion using sparse point trajectories via colored spline overlays on reference videos

  • vrgamedevgirl84

  • oumoumad

    • IC luminance map
    • LTX-2 IC-LoRA-Ungrade - Removes color grading and contrast from footage, returning neutral ungraded appearance
    • LTX-2.3 IC-LoRA-Ungrade - LTX-2.3 version of color grading removal IC-LoRA
    • IC-LoRA-Outpaint - Extends video canvas by generating new content in black regions (letterbox areas), filling with temporally consistent content
    • IC-LoRA-ReFocus - Removes lens blur and restores focus to out-of-focus footage (lens blur only)
    • IC-LoRA-Uncompress - Removes MP4 compression artifacts (blocking, banding, mosquito noise) and restores clean video
    • IC-LoRA-MotionDeblur - Removes motion blur from footage
    • IC-LoRA-Deinterlace - Removes interlacing artifacts from video
    • FXIC LTX2 IC-LoRA - Flux-inspired IC-LoRA for LTX video transformation with multiple optimizer variants (adamw, prodigy, masked) at various training steps
    • DeArchive LTX-2.3 - In-Context LoRA for restoring archive video (old B&W footage, low-res web rips, sepia-toned silent-era prints) into colored, high-definition modern cinematography (Rank 128, 5,000 steps)

  • Kijai

    • Realisdance - IC-LoRA trained on the realisdance dance set (312 MB)
    • SAM3D Body v42 - IC-LoRA tied to SAM3D body pose conditioning (624 MB)

  • Cseti

    • IC-LoRA-Cameraman v1 - Transfers camera movements (zoom, pan, tilt, orbit) from reference video to generated output
    • IC-LoRA-EditRefVid v1 - Edit reference video IC-LoRA for editing existing videos using reference guidance
    • IC-LoRA-Cameraman v2 - v2 of the Cameraman IC-LoRA with a larger and more diverse dataset; replicates camera motion from a reference video. No trigger word required.
    • IC-LoRA-CrossView Prompt v0.9 - Virtual second camera IC-LoRA: give it a reference video + a short camera-angle prompt using the trigger crossview. and re-render the same scene from a new viewpoint. v0.9 proof-of-concept trained on synthetic multi-view data; no starting image needed.
    • IC-LoRA-CrossView Warp v0.9 - Depth-warp novel-view IC-LoRA for LTX-Video 2.3 (22B). Given a video + a depth-warp of the same video (from the CrossViewWarp ComfyUI node, Depth Anything V2 input) and an azimuth/elevation/distance offset, renders the scene from that new viewpoint. Sibling of CrossView-Prompt v0.9. (192 MB)

  • 100percentrobot

    • Audio-Reactive LORA - Generates audio-reactive videos with motion synchronized to musical elements (beats, rhythm)

  • LiconStudio

    • VBVR-lora-I2V - Enhances video generation for complex reasoning tasks including multi-object interactions, physical causality, and spatial relationships
    • VBVR-lora-I2V Special
    • Licon MSR V2 - Multiple Subject Reference v2: IC-LoRA preserving character identity / clothing / objects across frames in multi-reference video generation. Improves consistency, stability and scene logic vs. v1. Requires the ComfyUI-Licon-MSR plugin.

  • TheBurgstall

    • LTX-2.3-Skin-Hair - Refines skin texture and hair rendering, reduces plastic skin artifacts, improves specular highlights
    • VR-360-Outpaint IC-LoRA - Outpaints standard widescreen footage into a full 360° equirectangular projection for immersive/VR viewing.
    • Seamless-Equirectangular LTX-2.3 LoRA - Rank-128 LoRA for equirectangular 360° text-to-video generation with LTX-2.3 (15k steps). Trigger Equirectangular. Pairs with the ComfyUI-Seamless-Equirectangular node pack and EquiRoPE / Geometric CFG / per-step roll / circular VAE / wrapped noise setup.

  • Nightfury16

  • siraxe

    • MergeGreen IC-lora - Maintains motion at start/end frames, use middle frames with RGB 0,191,0 (75% green fill) in IC-LoRA workflow
    • TTM IC-lora - Makes cutouts cartoony and adds cartoony characters to video scenes, based on the TTM approach (use with Img To Video bypass + Add Video IC-LoRA Guide node)

  • Lightricks LTX-2

    • Canny Control - Edge detection control for structural guidance
    • Depth Control - Depth map conditioning for 3D spatial control
    • Detailer - Enhances fine details and textures in generated videos
    • Pose Control - Human pose estimation control for motion guidance

Upscaler LoRAs:

  • LTX 2.3 Upscale IC-LoRA by Zlikwid
    • Generative refinement LoRA for upscaling lower-res or soft videos
    • Works by bicubic upscaling first, then running through LTX 2.3 with this LoRA
    • Use prompt: upscale
  • LTX2.3-ICEdit-Insight by JoyFox Lab
    • Task-aware video restoration and editing model family
    • Supports: Video Restoration, HD Enhancement, Watermark Removal, Subtitle Removal
  • Singularity LTX-2.3 OmniCine by WarmBloodAban
    • Comprehensive optimizer for LTX2.3 I2V and First/Last Frame workflows
    • Features: Limb Evolution, Shot Injection, Natural Expression, Physical Integrity, Cross-Style Potential
    • Uses "Singularity" prompting framework with 7-block bilingual structure
  • yuvraj108c
    • LTX-2.3-22b-IC-LoRA-Any-Trajectory-Instruction - IC-LoRA porting Any Trajectory Instruction (ATI) to LTX-2.3; lets users draw motion paths on an input image and have the model generate videos following those trajectories (splines). Trained on 25 video pairs at 768x768x81 bucket size, 3000 steps.
  • zghhui
    • OmniNFT RL-LoRA - Modality-wise Omni Diffusion Negative-aware Fine-Tuning RL-LoRA for joint audio-video generation (paper arXiv:2605.12480). Provides both LTX-2 and LTX-2.3 variants.
    • LTX-2.3-OmniNFT-RL-LoRA (Kijai mirror) - Same OmniNFT RL-LoRA repackaged by Kijai in bf16 (588 MB).
  • VetoBugger
    • LTX2.3-CrispLora - Crisp enhancement LoRA for LTX-2.3 (LTX2.3_Crisp_Enhance.safetensors).
  • SyFeee
    • LTX-2.3 Dual-Character LoRA - Image-to-video character-consistency LoRA tuned for two-character dialogue scenes and multi-shot cinematic generation. Works for ancient Chinese fantasy, modern urban, and 3D anime styles. Recommended strength 0.7-0.9 standalone, 0.3-0.5 when stacked with style LoRAs.
  • WarmBloodAban
    • Singularity LTX-2.3 OmniCine V1 - Updated V1 of the OmniCine integrated optimization framework. Restructures LTX-Video 2.3 generation logic with focus on I2V, First/Last Frame, and Reference-to-Video workflows. Nearly 100,000 training steps. Includes Singularity-LTX-2.3_OmniCine_V1.safetensors (2.57 GB) and Singularity-LTX-2.3_OmniCine_V1nsf.safetensors (2.57 GB, NSF variant).

▣ Styles

▣ Special

  • Wan2.1 VAE Adapter

    • Latent space adapter for converting between LTX-2 and Wan2.1 VAE representations
    • latent_adapter_final.pt (447 MB)

  • TenStrip

    • LTX2.3 JoyAI LoRA Extracted - LoRA extracted from jdopensource/JoyAI-Echo; boosts prompt response and motion in LTX-2.3 (also used for NSFW/Sulphur/Eros) at strength 0.4–0.7.
    • DMD LoRA (r256) - DMD-distillation delta extraction from JoyAI-Echo, reshaped for rank-256 sampling. Use at 1.0 with 8/4-step upscale or experiment with other sigmas; any euler or LTX-compatible sampler. No custom loading needed. (4.86 GB)
    • DMD LoRA Hybrid v1 - Hybrid DMD distillation: works inside blocks 0-25 to increase movement strength above the standalone DMD without letting the LTX distilled LoRA redraw or bring negative base-model tendencies. Improves movement smoothness and consistency. (4.86 GB)
    • DMD LoRA Hybrid v2 - v2 of the hybrid DMD distillation. (4.86 GB)

  • ltx-community

  • zzmicer

    • Sax - Saxophone audio LoRA (video-only, no-audio conditioning) at 4000 steps. (403 MB)
    • Violin (8s) - Violin audio LoRA trained on 8-second clips at 3000 steps. (856 MB)
    • Guitar - Acoustic guitar audio LoRA at 3000 steps. (1.71 GB)
    • DJ - DJ/electronic-music audio LoRA (rank 128, LR 5e-4) at 1800 steps. (1.71 GB)

  • Lightricks

    • LTX-2.3-22b-LoRA-Foley-V2A - Official Lightricks Foley V2A (video-to-audio) LoRA for LTX-2.3. Generates realistic, visually-synced Foley sound effects from video. Rank ~small (216 MB); pairs with the workflow JSON in the repo (ltx-2.3-foley-v2a.json).
    • LTX-2.3-22b-IC-LoRA-Clean-Plate - Official Lightricks Clean-Plate IC-LoRA. Removes people/objects from video frames for steady background plates. Rank 128 (312 MB); pairs with the LTX-2.3_-_V2V_mass_remove_people_Clean-Plate-Lora.json workflow.
    • LTX-2.3-22b-IC-LoRA-Relight - Official Lightricks Relight IC-LoRA (single-stage, distilled). Relights a scene using a reference sphere image that defines the target lighting direction and color. Pairs with the bundled LTX-2.3_Relight_ICLoRA_SingleStage_Distilled.json workflow. (312 MB)
    • LTX-2.3-22b-IC-LoRA-Cross-Eyed - Official Lightricks Cross-Eyed IC-LoRA v0.9. Converts standard video into side-by-side 3D stereo output (cross-eyed viewing). (327 MB)
    • LTX-2.3-22b-IC-LoRA-Colorization - Official Lightricks Colorization IC-LoRA v0.9. Adds realistic color to B&W or desaturated input video. (906 MB)
    • LTX-2.3-22b-IC-LoRA-Deblur - Official Lightricks Deblur IC-LoRA v0.9. Removes blur and restores sharp, focused video. (906 MB)
    • LTX-2.3-22b-IC-LoRA-Decompression - Official Lightricks Decompression IC-LoRA v0.9. Removes MP4 compression artifacts (banding, mosquito noise) and restores clean video. (906 MB)
    • LTX-2.3-22b-IC-LoRA-Water-Simulation - Official Lightricks Water-Simulation IC-LoRA v0.9. Adds physically-plausible fluid/water dynamics to reference video (reflections, ripples, splashes). (906 MB)
    • LTX-2.3-22b-IC-LoRA-Instant-Shave - Official Lightricks Instant-Shave IC-LoRA v0.9. Removes facial hair in video-to-video. (654 MB)
    • LTX-2.3-22b-IC-LoRA-In-Outpainting - Official Lightricks In-Outpainting IC-LoRA v0.9. Extends canvas in any direction from reference video, filling new regions with consistent scene content. (1.31 GB)
    • LTX-2.3-22b-IC-LoRA-Day-To-Night - Official Lightricks Day-To-Night IC-LoRA v0.9. Converts daytime footage to nighttime with matched lighting/shadows. (327 MB)
    • LTX-2.5-22b-IC-LoRA-Pixel-Spatial-Upscaler - Official Lightricks Pixel-Spatial-Upscaler IC-LoRA for LTX-2.5. Adds x2 pixel-level spatial upscaling to a video pass. (0.30 GB)

  • fbjr

    • LTX-2.3-22b-IC-LoRA-Audio-Only-Context - Two IC-LoRA checkpoints for audio-only and cross-modal audio-video context conditioning on LTX-2.3. Audio-only checkpoint (156 MB, audio_only_step_01000.safetensors) preserves a reference audio identity through a generation pass; cross-modal checkpoint (276 MB, cross_modal_step_01000.safetensors) extends that to joint audio-video conditioning.

  • FuzzPuppy

    • LTX-2.3 Foley - Video-to-audio LoRA for LTX-2.3 that adds realistic, visually synchronized Foley / sound effects over a video (multiplier 1.0–3.0; pairs with the LTX Community License).
    • Black Magic - Video-to-video IC-LoRA for low-light / shadow reconstruction. Restores underexposed footage to natural, well-exposed video with realistic colors and recovered shadow detail. Pairs with the bundled ltx23-black-magic-lora-workflow.json. (312 MB)

  • Hoffm4nz

    • LTX-2.3-22b-IC-LoRA-Golden-Hour - IC-LoRA fine-tune that biases generation toward warm "Golden Hour" lighting aesthetics. Root canonical checkpoint ltx-2.3-22b-ic-lora-golden-hour.safetensors (312 MB); four per-step snapshots are also available in the checkpoints/ folder of the repo (step_00250 → step_01000, 312 MB each).

  • JanKanta

    • LensRemover - Video-to-video IC-LoRA that removes lens flares, veiling glare, lens dirt, and other optical artifacts from footage while keeping the underlying scene intact. Dual-encoding: works on both sRGB/Rec.709 and ARRI LogC3 inputs (selected via trigger word). Pairs with the bundled LensRemover_comfyui_workflow.json. (312 MB)

  • vpakarinen

  • Comfy-Org

    • ltx-2.3-22b-ic-lora-ingredients-0.9 - Comfy-Org-distributed ingredients IC-LoRA for LTX-2.3 (the other three siblings in the same folder — id-lora-celebvhq-3k, id-lora-talkvid-3k, distilled_1.1_lora_dynamic_fro09_avg_rank_111_bf16 — are already linked from the Checkpoints / ID-LoRA tables).

  • linoyts

    • ltx2.3-ic-lora-ingredients-multishot - IC-LoRA fine-tune from ltx-2.3-22b-dev.safetensors (2,500 training steps, LR 1e-4). Trained for multi-shot ingredient-conditioned generation.
    • ltx2-ic-lora-ui - LoRA fine-tune from ltx-2.3-22b-dev.safetensors (20 training steps; treats UI-screen aesthetics). Inherits LTX-2.3 base license.

  • rzgar

    • Motion Enhancer (N54W) - General-purpose N54W motion enhancer LoRA. Enhances fluidity, coherence, and motion when stacked as a companion LoRA alongside other specialized models. Extensively tested with top-rated Civitai LoRAs. (2.17 GB)
    • Distilled LoRA 384-1.1 "n4w" — Modified official 384-1.1 distilled LoRA that cooperates with N54W LoRAs (instead of fighting them), improving motion and appearance. Two variants: full rank ~7.08 GB (ltx-2.3-22b-distilled-lora-384-1.1_n4w.safetensors) and low-VRAM rank-128 ~2.40 GB (ltx-2.3-22b-distilled-lora-384-1.1_rank_128_n4w.safetensors). Use at strength 0.5–0.65 stacked with your N54W LoRA. Apache-2.0.

  • joeygambino

    • American Accent (audio) - Audio-branch LoRA (rank 32) that makes accent wording in prompts actually work. LTX-2.3's voice prior ignores accent requests in certain regions (e.g. young female characters default to Australian); this LoRA turns accent prompts into reliable control. Requires 24 fps — off-24 fps overrides accent wording entirely. (168 MB)

▣ ID-LoRA (Identity-Driven In-Context LoRA)

ID-LoRA is a method that enables identity-preserving audio-video generation in a single model. It jointly generates a subject's appearance and voice, letting a text prompt, a reference image, and a short audio clip govern both modalities together. Built on top of LTX-2.3 (22B), it is the first method to personalize visual appearance and voice within a single generative pass.

Unlike cascaded pipelines that treat audio and video separately, ID-LoRA operates in a unified latent space where a single text prompt can simultaneously dictate the scene's visual content, environmental acoustics, and speaking style—while preserving the subject's vocal identity and visual likeness.

Key Features:

  • Text prompt controls the scene and content
  • Reference image preserves the subject's visual likeness
  • Short audio clip preserves the subject's vocal identity
  • Single unified generation pass for both appearance and voice

Available LoRAs for LTX-2.3:

LoRALoRA RankSizeDownload
ID-LoRA-TalkVid-3K1281.1 GB ┊
ID-LoRA-CelebVHQ-3K1281.1 GB ┊

Resources:

◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆

▓ ComfyUI Nodes

▣ Custom Node Collections

  • 10S-Comfy-nodes by TenStrip - Custom ComfyUI nodes for improving motion quality when working with LTX 2.3's combined audio/video latent pipeline. Includes Latent Cross Fade Auto Concat, Audio Latent Stretch, Latent Motion Sharpener, Latent Temporal Upsampler, Latent Motion Retime, and Latent Temporal Inpainter for clean 30fps output from 24fps sampled models.

  • Deno Custom Nodes by Deno2026 - Practical ComfyUI custom nodes focused on fast real-world workflow improvements including (Deno) Resize Box, Multi Image Loader, LTX Sequencer, LTX Model Loader, Easy Model Download Helper, LTX Multi LoRA Loader, and LTX Prompt Guide.

  • PromptRelay by kijai - Enables consistent multilingual lip-sync while maintaining voice consistency across languages. Distributes video latent frames across segments with smart prompt node supporting inline and block syntax styles.

  • WhatDreamsCost ComfyUI by WhatDreamsCost - LTX Director 2.0 plus a variety of custom ComfyUI nodes and workflows for creating AI-generated video content including Multi Image Loader, LTX Sequencer, LTX Keyframer, Speech Length Calculator, Load Video UI, and Load Audio UI. 1.7k+ stars.

  • ComfyUI-Sapiens2 by kijai - ComfyUI nodes for Sapiens2 computer vision models from Facebook Research. Supports pose estimation, body-part segmentation, surface normal estimation, and pointmap estimation with model variants from 400M to 5B parameters.

◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆

▓ LoRA Training

For training LTX LoRAs, the community uses a variety of official scripts, community-developed forks, and cloud-based platforms.

Primary Local Training Tools

  • Official LTX-2 Trainer: This is the standard Python-based package for training LoRAs, full fine-tuning, and In-Context (IC) LoRAs. It is designed for Linux and requires CUDA and Triton.
  • Musubi-Tuner (AkaneTendo25 Fork): Widely considered the fastest and most efficient local trainer for LTX-2 and 2.3. It features significantly smaller cache sizes (up to 12x smaller than AI Toolkit) and better iteration speeds, reaching up to 2 iterations per second on an RTX 5090.
  • AI Toolkit (by Ostris): A popular third-party tool that supports LTX-2 character and image-to-video LoRAs. While beginner-friendly, some users reported issues with audio training on the main branch.
  • AI Toolkit: BIG-DADDY-VERSION (ArtDesignAwesome Fork): This specific fork was created to fix broken audio and voice training in the original AI Toolkit. It is optimized for hardware like the RTX 5090.
  • rs-nodes (richservo): A collection of nodes that includes a full LTX Lora trainer directly within ComfyUI. It is designed to be memory-efficient, allowing training on cards with as little as 11GB-12GB of VRAM by using ComfyUI's native weight loaders.
  • SimpleTuner: A highly optimized trainer for Linux that supports LTX-2 and is noted for its ability to handle larger datasets on limited VRAM via block swapping.

Cloud Training Platforms

  • Fal.ai: Provides a dedicated cloud trainer for custom styles and effects, though it is primarily limited to image-based training datasets.
  • RunComfy: A cloud service that offers a pre-configured AI Toolkit setup specifically for LTX-2 training.

Essential Dataset & Captioning Tools

  • Taz's Ultimate Captioning Tool: A Hugging Face space frequently used by the community to generate the long, detailed, cinematographic prompts (around 200 words) that LTX-2 requires for high-quality training.
  • AI Video Clipper & LoRA Captioner: A modular pipeline designed to automate local dataset creation using WhisperX and Qwen2-VL, including support for RTX 5090 Blackwell cards.

Training Requirements Summary

  • Dataset: Videos should typically be cut to 121 frames (exactly 4.84 seconds) to align with the model's architectural "8n+1" rule.
  • Hardware: While 16GB VRAM is possible with extreme offloading in tools like rs-nodes, 24GB is the practical minimum for quantized training. For best results and speed, 48GB to 80GB (H100 or RTX 6000) is preferred.
  • Precision: It is now officially recommended to train on the full BF16 model for LTX 2.3 rather than FP8 for superior quality.

◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆◇◆

▓ Workflow & Technical Notes

❖ RuneXX

RuneXX LTX-2.3 Workflows:

Movie-Maker:

Multi-ref-character-sheet:

Helper-wf:

Talking-Avatar-TTS:

Video-2-Video:

Just-Talk — add voice to silent video:

Extend-Any-Video:

Inpaint:

Shot-to-Shot-Transition:

Other:

**Music-Video-Creator:**Music-Video-Creator:

Others:

Custom-Audio:

First-Last-Frame:

Long-Video-Experimental:

3-Pass-Experimental:

Control-reference:

Helper-Workflows:

Other-examples:

RuneXX LTX-2 Workflows old pre_feb2026

═══●═══●═══●═══●═══●═══●═══●═══●═══

❖ Lightricks

LTX-2.3:

LTX-2:

❖ vrgamedevgirl84

vrgamedevgirl84 LTX 2.3 Music Video Creator:

  • Music Video Creator Workflow
    • Prompt Creator Workflow - Audio upload, beat detection, scene timing, lyrics analysis, style selection, prompt generation
    • Text-to-Video Workflow - LoRA integration, advanced prompt controls, Remake Mode, video stitching
    • Image-to-Video Workflow - Uses Z-Image Turbo and LTX 2.3
    • Requirements: ComfyUI, LTX 2.3 models, Z-Image Turbo model, FFmpeg, vrgamedevgirl custom nodes

❖ ComfyUI

关于 About

All available LTX-2 models, encoders, workflows, LoRAs for ComfyUI
aicomfyuiitvloraltx-2ltxvmodelt2vvideovideo-ai

语言 Languages

提交活跃度 Commit Activity

代码提交热力图
过去 52 周的开发活跃度
116
Total Commits
峰值: 23次/周
Less
More

核心贡献者 Contributors