LTX-2.5 (22B) VRAM requirements
The July 2026 LTX, with a custom Gemma 4 12B text encoder. GGUF sizes differ a lot between uploaders because they mix precisions differently; figures here use realrebelai's distilled files.
TypeVideo
Parameters22B
Fits entirely from14 GB
8-bit or better from29 GB
01The files, and how much VRAM each needs
| File | Size | Needed | Min. VRAM | Quality | Source |
|---|---|---|---|---|---|
| 16-bit | 42.0 GB | 46.8 GB | 47 GB | the original weights | Lightricks/LTX-2.5 → |
| Q8_0 | 23.6 GB | 28.4 GB | 29 GB | practically identical to the original | realrebelai/LTX-2.5_GGUFs → |
| Q6_K | 18.7 GB | 23.5 GB | 24 GB | very close to the original | realrebelai/LTX-2.5_GGUFs → |
| Q5_K_M | 16.8 GB | 21.6 GB | 22 GB | close; small differences in fine detail | realrebelai/LTX-2.5_GGUFs → |
| Q4_K_M | 15.1 GB | 19.9 GB | 20 GB | good; some loss in fine detail and text | realrebelai/LTX-2.5_GGUFs → |
| Q3_K_M | 11.5 GB | 16.3 GB | 17 GB | noticeable loss of detail | realrebelai/LTX-2.5_GGUFs → |
| Q2_K | 8.8 GB | 13.6 GB | 14 GB | heavy loss; a last resort | realrebelai/LTX-2.5_GGUFs → |
“Needed” = file + 4 GB working memory + 0.8 GB system reserve.
02By amount of VRAM
03Best GPU for LTX-2.5
The cheapest cards (by launch price) that run it well, and every card sorted by memory: best GPU for LTX-2.5 → Planning bigger images or longer clips? Open the calculator →
04By graphics card
| GPU | VRAM | Verdict | Best file | Needed |
|---|---|---|---|---|
| Desktop graphics cards | ||||
| RTX 2060 6 GB | 6 GB | Not practical | Q2_K | 13.6 GB |
| RTX 3050 6 GB | 6 GB | Not practical | Q2_K | 13.6 GB |
| RTX 2070 Super 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 2080 Super 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 3050 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 3060 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 3060 Ti 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 3070 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 3070 Ti 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 4060 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 4060 Ti 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 5050 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 5060 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 5060 Ti 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RX 7600 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RX 9050 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RX 9060 XT 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| Arc B570 10 GB | 10 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 3080 10 GB | 10 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 2080 Ti 11 GB | 11 GB | Offload only | Q4_K_M | 19.9 GB |
| Arc B580 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RTX 2060 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RTX 3060 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RTX 3080 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RTX 3080 Ti 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RTX 4070 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RTX 4070 Super 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RTX 4070 Ti 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RTX 5070 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RX 7700 XT 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RX 9070 GRE 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| Arc A770 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RTX 4060 Ti 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RTX 4070 Ti Super 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RTX 4080 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RTX 4080 Super 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RTX 5060 Ti 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RTX 5070 Ti 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RTX 5080 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RX 7600 XT 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RX 7800 XT 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RX 7900 GRE 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RX 9060 XT 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RX 9070 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RX 9070 XT 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RX 7900 XT 20 GB | 20 GB | Runs | Q4_K_M | 19.9 GB |
| Arc Pro B60 24 GB | 24 GB | Runs | Q6_K | 23.5 GB |
| RTX 3090 24 GB | 24 GB | Runs | Q6_K | 23.5 GB |
| RTX 3090 Ti 24 GB | 24 GB | Runs | Q6_K | 23.5 GB |
| RTX 4090 24 GB | 24 GB | Runs | Q6_K | 23.5 GB |
| RX 7900 XTX 24 GB | 24 GB | Runs | Q6_K | 23.5 GB |
| Arc Pro B70 32 GB | 32 GB | Runs well | Q8_0 | 28.4 GB |
| RTX 5090 32 GB | 32 GB | Runs well | Q8_0 | 28.4 GB |
| Laptop GPUs | ||||
| RTX 3050 Laptop 4 GB | 4 GB | Not practical | Q2_K | 13.6 GB |
| RTX 3050 Ti Laptop 4 GB | 4 GB | Not practical | Q2_K | 13.6 GB |
| RTX 2060 Laptop 6 GB | 6 GB | Not practical | Q2_K | 13.6 GB |
| RTX 3050 Laptop 6 GB | 6 GB | Not practical | Q2_K | 13.6 GB |
| RTX 3060 Laptop 6 GB | 6 GB | Not practical | Q2_K | 13.6 GB |
| RTX 4050 Laptop 6 GB | 6 GB | Not practical | Q2_K | 13.6 GB |
| RTX 2070 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 2070 Super Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 2080 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 2080 Super Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 3070 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 3070 Ti Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 3080 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 4060 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 4070 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 5050 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 5060 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 5070 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RX 7600M 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RX 7600M XT 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RX 7600S 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RX 7700S 8 GB | 8 GB | Offload only | Q4_K_M | 19.9 GB |
| RTX 4080 Laptop 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RTX 5070 Laptop 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RTX 5070 Ti Laptop 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RX 7800M 12 GB | 12 GB | Offload only | Q2_K | 13.6 GB |
| RTX 3080 Laptop 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RTX 3080 Ti Laptop 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RTX 4090 Laptop 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RTX 5080 Laptop 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RX 7900M 16 GB | 16 GB | Tight | Q2_K | 13.6 GB |
| RTX 5090 Laptop 24 GB | 24 GB | Runs | Q6_K | 23.5 GB |
| Unified memory | ||||
| Radeon 8060S (Strix Halo) 96 GB | 96 GB | Runs well | 16-bit | 46.8 GB |
| Radeon 8065S (Gorgon Halo) 160 GB | 160 GB | Runs well | 16-bit | 46.8 GB |
| Apple Silicon Macs (by memory) | ||||
| Mac 16 GB | 12.7 GB | Offload only | Q2_K | 13.6 GB |
| Mac 18 GB | 14.4 GB | Tight | Q2_K | 13.6 GB |
| Mac 24 GB | 19.6 GB | Tight | Q3_K_M | 16.3 GB |
| Mac 32 GB | 26.8 GB | Runs | Q6_K | 23.5 GB |
| Mac 36 GB | 30.2 GB | Runs well | Q8_0 | 28.4 GB |
| Mac 48 GB | 40.2 GB | Runs well | Q8_0 | 28.4 GB |
| Mac 64 GB | 55.7 GB | Runs well | 16-bit | 46.8 GB |
| Mac 96 GB | 85 GB | Runs well | 16-bit | 46.8 GB |
| Mac 128 GB | 115.4 GB | Runs well | 16-bit | 46.8 GB |
| Mac 192 GB | 175.4 GB | Runs well | 16-bit | 46.8 GB |
| Mac 256 GB | 236.9 GB | Runs well | 16-bit | 46.8 GB |
| Mac 512 GB | 498.1 GB | Runs well | 16-bit | 46.8 GB |
05Text encoder, VAE and other files
| File | Folder | Size | When | |
|---|---|---|---|---|
| Gemma 4 12B + LTX projection INT8 (convrot) gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors | models/text_encoders | 15.4 GB | text encoder · smaller, recommended | Download → |
| Gemma 4 12B + LTX projection BF16 gemma4-12b-with-proj-ltx-2.5-bf16.safetensors | models/text_encoders | 26.3 GB | text encoder · alternative | Download → |
| LTX-2.5 video VAE (DiffVAE) BF16 ltx-2.5-video-vae-bf16.safetensors | models/vae | 1.5 GB | required | Download → |
| LTX-2.5 audio VAE + vocoder BF16 ltx-2.5-audio-vae-bf16.safetensors | models/vae | 0.4 GB | required | Download → |
| LTX-2.5 latent spatial upscaler x2 ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors | models/latent_upscale_models | 1.0 GB | optional · official two-stage template | Download → |
| Gemma 4 E2B INT8 (prompt enhancer, optional) gemma4_e2b_it_int8_convrot.safetensors | models/text_encoders | 5.2 GB | optional · only if prompt enhancer enabled | Download → |
The files the official ComfyUI workflows load next to the model. Sizes read from Hugging Face (2026-10-01).
At 15.4 GB even the smallest Gemma 4 12B (custom LTX-2.5 build) file is too big for 16 GB, so it runs from system RAM. That slows down reading the prompt, not the picture. Sizes: 16-bit 26.3 GB, INT8 15.4 GB. Why the encoder does not count against VRAM →
06Where the files go in ComfyUI
| File | Folder | Loader node |
|---|---|---|
| Diffusion model (.safetensors: 16-bit, FP8, INT8) | ComfyUI/models/diffusion_models | Load Diffusion Model |
| GGUF file (.gguf) | ComfyUI/models/unet | Unet Loader (GGUF) - from the ComfyUI-GGUF node pack |
| Text encoder | ComfyUI/models/text_encoders | Load CLIP / DualCLIPLoader (or the GGUF versions) |
| VAE | ComfyUI/models/vae | Load VAE |
07Measured and reported results
No measured results yet. Send yours.
08Training a LoRA for LTX-2.5
No trainer documentation with a VRAM figure for this model was found yet. What the trainers say for other models →
09Every file tracked for LTX-2.5
| File | Type | Size | Repo |
|---|---|---|---|
| LTX-2.5-Distilled-Q8_0.gguf (distilled) | Q8_0 | 23.60 GB | Abiray/LTX-2.5-Distilled-GGUF → |
| LTX-2.5-Distilled-Q6_K.gguf (distilled) | Q6_K | 18.62 GB | Abiray/LTX-2.5-Distilled-GGUF → |
| LTX-2.5-Distilled-Q5_K_M.gguf (distilled) | Q5_K_M | 18.12 GB | Abiray/LTX-2.5-Distilled-GGUF → |
| LTX-2.5-Distilled-Q4_K_M.gguf (distilled) | Q4_K_M | 15.69 GB | Abiray/LTX-2.5-Distilled-GGUF → |
| LTX-2.5-Distilled-Q4_K_S.gguf (distilled) | Q4_K_S | 15.33 GB | Abiray/LTX-2.5-Distilled-GGUF → |
| LTX-2.5-Distilled-Q3_K_M.gguf (distilled) | Q3_K_M | 12.92 GB | Abiray/LTX-2.5-Distilled-GGUF → |
| LTX-2.5-Distilled-Q3_K_S.gguf (distilled) | Q3_K_S | 12.65 GB | Abiray/LTX-2.5-Distilled-GGUF → |
| ltx-2.5-22b-dev-transformer-bf16.safetensors (dev) | BF16 | 42.02 GB | Lightricks/LTX-2.5 → |
| ltx-2.5-22b-distilled-transformer-bf16.safetensors (distilled) | BF16 | 42.02 GB | Lightricks/LTX-2.5 → |
| ltx-2.5-22b-dev-transformer-comfy-int8-convrot.safetensors (dev) | INT8 | 21.50 GB | Lightricks/LTX-2.5 → |
| ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (distilled) | INT8 | 21.50 GB | Lightricks/LTX-2.5 → |
| ltx-2.5-22b-distilled-transformer-nvfp4.safetensors (distilled) | NVFP4 | 18.72 GB | Lightricks/LTX-2.5 → |
| LTX-2.5-Distilled-Q8_0.gguf (distilled) | Q8_0 | 23.63 GB | realrebelai/LTX-2.5_GGUFs → |
| LTX-2.5-Distilled-Q6_K.gguf (distilled) | Q6_K | 18.66 GB | realrebelai/LTX-2.5_GGUFs → |
| LTX-2.5-Distilled-Q5_K_M.gguf (distilled) | Q5_K_M | 16.82 GB | realrebelai/LTX-2.5_GGUFs → |
| LTX-2.5-Distilled-Q4_K_M.gguf (distilled) | Q4_K_M | 15.09 GB | realrebelai/LTX-2.5_GGUFs → |
| LTX-2.5-Distilled-Q4_K_S.gguf (distilled) | Q4_K_S | 13.85 GB | realrebelai/LTX-2.5_GGUFs → |
| LTX-2.5-Distilled-Q3_K_M.gguf (distilled) | Q3_K_M | 11.53 GB | realrebelai/LTX-2.5_GGUFs → |
| LTX-2.5-Distilled-Q2_K.gguf (distilled) | Q2_K | 8.83 GB | realrebelai/LTX-2.5_GGUFs → |
| ltx-2.5-22b-dev-transformer-Q8_0.gguf (dev) | Q8_0 | 23.60 GB | vantagewithai/LTX-2.5-GGUF → |
| ltx-2.5-22b-dev-transformer-Q6_K.gguf (dev) | Q6_K | 18.62 GB | vantagewithai/LTX-2.5-GGUF → |
| ltx-2.5-22b-dev-transformer-Q5_K_M.gguf (dev) | Q5_K_M | 18.12 GB | vantagewithai/LTX-2.5-GGUF → |
| ltx-2.5-22b-dev-transformer-Q5_1.gguf (dev) | Q5_1 | 16.26 GB | vantagewithai/LTX-2.5-GGUF → |
| ltx-2.5-22b-dev-transformer-Q5_0.gguf (dev) | Q5_0 | 15.98 GB | vantagewithai/LTX-2.5-GGUF → |
| ltx-2.5-22b-dev-transformer-Q5_K_S.gguf (dev) | Q5_K_S | 15.89 GB | vantagewithai/LTX-2.5-GGUF → |
| ltx-2.5-22b-dev-transformer-Q4_K_M.gguf (dev) | Q4_K_M | 15.69 GB | vantagewithai/LTX-2.5-GGUF → |
| ltx-2.5-22b-dev-transformer-Q4_1.gguf (dev) | Q4_1 | 15.53 GB | vantagewithai/LTX-2.5-GGUF → |
| ltx-2.5-22b-dev-transformer-Q4_K_S.gguf (dev) | Q4_K_S | 15.33 GB | vantagewithai/LTX-2.5-GGUF → |
| ltx-2.5-22b-dev-transformer-Q4_0.gguf (dev) | Q4_0 | 15.24 GB | vantagewithai/LTX-2.5-GGUF → |
| ltx-2.5-22b-dev-transformer-Q3_K_M.gguf (dev) | Q3_K_M | 12.92 GB | vantagewithai/LTX-2.5-GGUF → |
| ltx-2.5-22b-dev-transformer-Q3_K_S.gguf (dev) | Q3_K_S | 12.65 GB | vantagewithai/LTX-2.5-GGUF → |
| ltx-2.5-22b-dev-transformer-Q2_K.gguf (dev) | Q2_K | 12.13 GB | vantagewithai/LTX-2.5-GGUF → |
EXTRA (Jul 2026, newest LTX; Lightricks repo ~1.6M downloads). GGUF sizes differ a lot between repos because of different mixed-precision recipes - pick repo carefully. Text encoder is a Gemma model (not fetched).