ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors
LTX 2.5 Distilled INT8 convrot (Official)
Official INT8 convrot quantized distilled transformer for LTX 2.5 — the first-party ComfyUI quant. Fastest official 2.5 inference path. Pairs with a ~15.37GB+ Gemma 4 text encoder, so a fully-official pipeline needs well above 24GB. Repo is gated.
Download ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors
Official HuggingFace repo — sign in and click "Agree and Access" to accept the LTX 2.5 license before this downloads. 21.50 GB · Free.
No 24GB GPU? Try ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors online — free generation included
Skip the 21.50 GB download and ComfyUI setup. Generate a 6-second video using this exact model in your browser, ~30 seconds.
When to choose ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors
Best official starting point on 24GB+ cards. On 16GB, the official Gemma 4 encoder alone (15.37GB) plus this transformer already exceeds your VRAM — use the community GGUF distilled transformer + GGUF Gemma 4 encoder instead.
Will this run on my GPU?
Minimum: 24GB VRAM. Headroom up to: 32GB.
⚠ FP8 scaled matmul requires RTX 40-series or newer (Ada Lovelace architecture). RTX 30xx cannot run this format — use the MXFP8 block-32 or BF16 variant instead.
How to use ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors
- Download the file from HuggingFace.
- Place it in ComfyUI/models/checkpoints/ inside your ComfyUI directory.
- Restart ComfyUI (or refresh the model list from the menu).
- Load a compatible workflow — see below.
See the LTX 2.5 ComfyUI setup guide for the full file list and install order.
Don't want to run this locally? Try ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors online with a free generation — no GPU, no install, ~30 seconds per clip.
ComfyUI says it can't find ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors?
Some published workflow JSONs reference this file under a custom subdirectory. If ComfyUI shows a "cannot find model" error and your workflow references one of these path-prefixed variants:
- diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors
The prefix before the slash or backslash is a subdirectory the workflow author used. The actual file is the same ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors — you have two fixes:
- Create the matching subdirectory inside ComfyUI/models/checkpoints/ and place the file there. Example: if the workflow references diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors, create the corresponding subfolder under ComfyUI/models/checkpoints/ and put ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors inside it.
- Or open the workflow JSON in a text editor and replace the prefixed string with just ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors. ComfyUI then resolves it directly from ComfyUI/models/checkpoints/.
On Windows the separator is \, on macOS/Linux it is / — they refer to the same nested folder regardless of platform.
Common issues
ComfyUI doesn't see the file after I downloaded it▼
Make sure the file is in ComfyUI/models/checkpoints/ (not a subfolder). Restart ComfyUI fully — the menu refresh sometimes misses new files. Filename must match exactly: ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors.
I get a CUDA error mentioning fp8 / scaled / matmul▼
FP8 scaled matmuls require an RTX 40-series GPU or newer (Ada Lovelace architecture). RTX 30-series and older cannot run FP8 weights at native precision. Use the BF16 variant instead, or the MXFP8 block-32 alternative.
CUDA out of memory error when loading the model▼
ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors needs ~24GB VRAM minimum. If you're hitting OOM: • Enable Sequential Offloading in ComfyUI settings • Lower the resolution (768×512 instead of 1280×704) — both dimensions must be divisible by 32 • Reduce frame count (65 frames instead of 161) — must be 8n+1 • Use a smaller variant — see Related models below.
Get notified when LTX 2.5 Distilled INT8 convrot (Official) updates
Occasional updates on what's new in LTX 2.5 — new quants, LoRAs, releases — with our hands-on verdict on whether they're worth re-downloading. No fixed cadence.
No spam. Sent occasionally when there's real news. Unsubscribe in one click.