What Can the NVIDIA RTX 4070 Ti Run?
12 GB · compatibility with 168 AI models
Home › GPUs › NVIDIA RTX 4070 Ti
Short answer: the NVIDIA RTX 4070 Ti (12 GB) can run 165 of 168 AI image, video, 3D and audio models we track. Best picks: Qwen-Image 2.1, Z-Image Turbo, WAN 2.2 T2V A14B (MoE), WAN 2.2 TI2V 5B, WAN 2.2 I2V A14B.
| Perfect | Great | Good | Usable | Barely | Can't Run |
|---|---|---|---|---|---|
| 13 models | 64 models | 68 models | 20 models | 0 models | 3 models |
NVIDIA RTX 4070 Ti specs for AI
| Brand | NVIDIA |
|---|---|
| Series | RTX 40 |
| VRAM | 12 GB |
| Architecture | ada |
| FP16 compute | 40.1 TFLOPS |
| Memory bandwidth | 504 GB/s |
| FP8 support | Yes |
| Released | 2023 |
| Launch price | $699 |
AI models on the NVIDIA RTX 4070 Ti
Each model uses the best format that fits in 12 GB. How ratings work.
Text to Image
| Model | Rating | Runs as | VRAM used | Speed |
|---|---|---|---|---|
| Stable Diffusion 1.5 | Great | FP16 | 4 GB | Fast |
| Stable Diffusion 1.4 | Great | FP16 | 4 GB | Fast |
| Stable Diffusion 2.1 | Great | FP16 | 5 GB | Fast |
| Stable Diffusion 2.0 | Great | FP16 | 5 GB | Fast |
| PixArt-Alpha 600M | Great | FP16 | 6 GB | Fast |
| SDXL Base 1.0 | Great | FP16 | 7 GB | Fast |
| SDXL Turbo | Great | FP16 | 7 GB | Fast |
| Animagine XL 3.1 | Great | FP16 | 7 GB | Fast |
| Pony Diffusion v6 XL | Great | FP16 | 7 GB | Fast |
| NoobAI XL | Great | FP16 | 7 GB | Fast |
| Juggernaut XL Ragnarok | Great | FP16 | 7 GB | Fast |
| SDXL Lightning 4-step | Great | FP16 | 7 GB | Fast |
| DreamShaper XL | Great | FP16 | 7 GB | Fast |
| Illustrious XL v1.0 | Great | FP16 | 7 GB | Fast |
| Playground v2.5 2.6B | Great | FP16 | 7 GB | Fast |
| RealVisXL v5.0 | Great | FP16 | 7 GB | Fast |
| Epicrealism XL | Great | FP16 | 7 GB | Fast |
| PixArt-Sigma 900M | Great | FP16 | 7 GB | Fast |
| Stable Diffusion 3 Medium (2B) | Great | FP16 | 7 GB | Fast |
| SDXL Hyper 1-step | Great | FP16 | 7 GB | Fast |
| Stable Diffusion 3.5 Medium (2.5B) | Great | FP16 | 8 GB | Fast |
| HunyuanDiT 1.5B | Great | FP16 | 8 GB | Fast |
| Qwen-Image 2.1 | Good | FP8 | 7 GB | Fast |
| Flux.2 Klein 4B | Good | FP8 | 8 GB | Fast |
| Z-Image Base | Good | FP8 | 8 GB | Fast |
| ERNIE-Image 8B | Good | GGUF Q4 | 5 GB | Moderate |
| HiDream-O1-Image 8.8B | Good | FP8 | 8 GB | Fast |
| Kolors | Good | FP8 | 8 GB | Fast |
| Ming-Image 0.1 Design | Good | GGUF Q4 | 6 GB | Moderate |
| Z-Image Turbo | Good | FP16 | 10 GB | Fast |
| FLUX.2 Klein 9B | Good | FP8 | 9 GB | Moderate |
| Chroma1 HD | Good | FP8 | 9 GB | Moderate |
| Krea 2 Turbo | Good | GGUF Q4 | 8 GB | Moderate |
| Stable Diffusion 3.5 Large (8B) | Good | FP8 | 10 GB | Moderate |
| Stable Diffusion 3.5 Large Turbo | Good | FP8 | 10 GB | Moderate |
| Ideogram 4.0 | Good | FP8 | 10 GB | Moderate |
| Seedream 3.0 | Good | FP8 | 10 GB | Moderate |
| GLM-Image | Good | FP8 | 10 GB | Moderate |
| MOSS-Image 5B | Good | FP16 | 12 GB | Fast |
| Flux.1 Lite | Good | FP16 | 12 GB | Fast |
| AuraFlow 0.3 | Good | FP16 | 12 GB | Fast |
| Flux.1 Dev 12B | Usable | FP8 | 12 GB | Moderate |
| Flux.1 Schnell 12B | Usable | FP8 | 12 GB | Moderate |
| FLUX.1 Krea Dev | Usable | FP8 | 12 GB | Moderate |
| HunyuanImage 3.0 | Usable | FP8 | 12 GB | Moderate |
| Qwen-Image 2512 (20B) | Usable | GGUF Q4 | 12 GB | Moderate |
| Flux.2 Dev 32B | Usable | Offload | 12 GB | Very Slow |
Image to Image
| Model | Rating | Runs as | VRAM used | Speed |
|---|---|---|---|---|
| SD 1.5 img2img | Great | FP16 | 4 GB | Fast |
| InstructPix2Pix | Great | FP16 | 5 GB | Fast |
| SDXL img2img | Great | FP16 | 7 GB | Fast |
| IP-Adapter SDXL | Good | FP16 | 9 GB | Fast |
| LongCat-Image Edit | Good | GGUF Q4 | 6 GB | Moderate |
| IP-Adapter FaceID SDXL | Good | FP16 | 10 GB | Fast |
| Mage-Flow 4B | Good | FP16 | 10 GB | Fast |
| Boogu-Image Edit | Good | FP8 | 11 GB | Moderate |
| Flux.1 Dev img2img 12B | Usable | FP8 | 12 GB | Moderate |
| Flux Kontext Dev | Usable | FP8 | 12 GB | Moderate |
| Qwen-Image-Edit 2511 | Usable | GGUF Q4 | 12 GB | Moderate |
Inpainting
| Model | Rating | Runs as | VRAM used | Speed |
|---|---|---|---|---|
| SD 1.5 Inpainting | Great | FP16 | 4 GB | Fast |
| SDXL Inpainting | Great | FP16 | 7 GB | Fast |
| PowerPaint v2.1 | Great | FP16 | 7 GB | Fast |
| BrushNet | Great | FP16 | 8 GB | Fast |
| Flux.1 Fill Dev | Usable | FP8 | 12 GB | Moderate |
ControlNet
| Model | Rating | Runs as | VRAM used | Speed |
|---|---|---|---|---|
| ControlNet SD 1.5 | Great | FP16 | 5 GB | Fast |
| T2I-Adapter SDXL | Great | FP16 | 8 GB | Fast |
| ControlNet SDXL | Good | FP16 | 9 GB | Fast |
| ControlNet Flux | Good | GGUF Q4 | 10 GB | Moderate |
Video
| Model | Rating | Runs as | VRAM used | Speed |
|---|---|---|---|---|
| WAN 2.1 T2V 1.3B | Great | FP16 | 6 GB | Fast |
| AnimateDiff Lightning | Great | FP16 | 6 GB | Fast |
| AnimateDiff v3 (SD 1.5) | Great | FP16 | 6 GB | Fast |
| ZeroScope v2 576w | Great | FP16 | 6 GB | Fast |
| AnimateDiff V2V | Great | FP16 | 6 GB | Fast |
| AnimateDiff I2V | Great | FP16 | 6 GB | Fast |
| Stable Video Diffusion 1.1 | Great | FP16 | 8 GB | Fast |
| CogVideoX 2B | Great | FP16 | 8 GB | Fast |
| ModelScope T2V | Great | FP16 | 8 GB | Fast |
| TokenFlow | Great | FP16 | 8 GB | Fast |
| Latte | Great | FP16 | 8 GB | Fast |
| SEINE | Great | FP16 | 8 GB | Fast |
| CogVideoX 5B | Good | FP8 | 8 GB | Fast |
| CogVideoX 5B I2V | Good | FP8 | 8 GB | Fast |
| SkyReels V1 | Good | FP8 | 8 GB | Fast |
| SkyReels V1 I2V | Good | FP8 | 8 GB | Fast |
| Kandinsky 6.0 Video Lite 3B | Good | FP16 | 9 GB | Fast |
| LTX-Video 0.9 (2B) | Good | FP16 | 10 GB | Fast |
| LTX-Video I2V | Good | FP16 | 10 GB | Fast |
| Stable Video Diffusion XT 1.1 (25 frames) | Good | FP16 | 10 GB | Fast |
| DynamiCrafter 512 | Good | FP16 | 10 GB | Fast |
| WAN 2.2 T2V A14B (MoE) | Good | GGUF Q4 | 8 GB | Moderate |
| WAN 2.2 I2V A14B | Good | GGUF Q4 | 8 GB | Moderate |
| WAN 2.1 T2V 14B | Good | GGUF Q4 | 8 GB | Moderate |
| WAN 2.1 I2V 480P 14B | Good | GGUF Q4 | 8 GB | Moderate |
| WAN 2.1 V2V | Good | GGUF Q4 | 8 GB | Moderate |
| HunyuanVideo T2V v1.5 | Good | GGUF Q4 | 8 GB | Moderate |
| HunyuanVideo I2V v1.5 | Good | GGUF Q4 | 8 GB | Moderate |
| HunyuanVideo 13B T2V | Good | GGUF Q4 | 8 GB | Moderate |
| HunyuanVideo I2V | Good | GGUF Q4 | 8 GB | Moderate |
| SkyReels V3 R2V 14B | Good | GGUF Q4 | 9 GB | Moderate |
| Wan 2.2 S2V 14B | Good | GGUF Q4 | 10 GB | Moderate |
| WAN 2.1 I2V 720P 14B | Good | GGUF Q4 | 10 GB | Moderate |
| Wan-Animate-2 14B | Good | GGUF Q4 | 10 GB | Moderate |
| SCAIL-2 14B | Good | GGUF Q4 | 10 GB | Moderate |
| Wan-Dancer 14B | Good | GGUF Q4 | 10 GB | Moderate |
| HuMo 17B | Good | GGUF Q4 | 10 GB | Moderate |
| LongCat-Video 13.6B | Good | Offload | 6 GB | Very Slow |
| Wan 2.2 Animate 14B | Good | GGUF Q4 | 11 GB | Moderate |
| LTX-2 19B | Good | GGUF Q4 | 11 GB | Moderate |
| FramePack I2V HY FP8 | Good | Offload | 8 GB | Very Slow |
| WAN 2.2 TI2V 5B | Good | FP16 | 12 GB | Fast |
| LTX-2.5 22B | Good | GGUF Q3 | 11 GB | Slow |
| LTX-2.3 22B | Good | GGUF Q3 | 11 GB | Slow |
| AnimateDiff SDXL | Good | FP16 | 12 GB | Fast |
| Waver 1.0 | Good | FP16 | 12 GB | Fast |
| Ovi 1.1 | Usable | FP8 | 12 GB | Moderate |
| MiniMax H3 FL2VA | Usable | GGUF Q4 | 12 GB | Moderate |
| MiniMax H3 Ref2VA | Usable | GGUF Q4 | 12 GB | Moderate |
| LTX-Video 13B | Usable | Offload | 12 GB | Very Slow |
| Open-Sora 2.0 (11B) | Usable | Offload | 12 GB | Very Slow |
| MAGI-1 | Usable | Offload | 12 GB | Very Slow |
| DynamiCrafter 1024 | Usable | Offload | 12 GB | Very Slow |
| MOVA | Can't Run | — | needs 29 GB | — |
| Mochi 1 10B | Can't Run | — | needs 16 GB | — |
Upscaler
| Model | Rating | Runs as | VRAM used | Speed |
|---|---|---|---|---|
| 4x-UltraSharp | Perfect | FP32 | 0.3 GB | Fast |
| 4x-Remacri | Perfect | FP32 | 0.3 GB | Fast |
| SwinIR | Perfect | FP32 | 0.5 GB | Fast |
| RealESRGAN x4plus | Perfect | FP32 | 0.3 GB | Fast |
| HAT | Perfect | FP32 | 0.5 GB | Fast |
| DAT | Perfect | FP32 | 0.5 GB | Fast |
| ScuNET | Perfect | FP32 | 0.3 GB | Fast |
| LDSR | Great | FP16 | 4 GB | Fast |
| SeedVR2 7B | Good | FP8 | 8 GB | Fast |
| SeedVR2 3B | Good | FP16 | 9 GB | Fast |
Face/Restore
| Model | Rating | Runs as | VRAM used | Speed |
|---|---|---|---|---|
| GFPGAN v1.4 | Perfect | FP32 | 0.3 GB | Fast |
| CodeFormer v0.1 | Perfect | FP32 | 0.3 GB | Fast |
VAE
| Model | Rating | Runs as | VRAM used | Speed |
|---|---|---|---|---|
| SD 1.5 VAE ft-mse | Perfect | FP32 | 0.2 GB | Fast |
| Flux VAE | Perfect | FP32 | 0.3 GB | Fast |
| SDXL VAE | Perfect | FP32 | 0.3 GB | Fast |
| TAESD / TAESDXL | Perfect | FP32 | 0.1 GB | Fast |
3D
| Model | Rating | Runs as | VRAM used | Speed |
|---|---|---|---|---|
| TripoSR | Great | FP16 | 4 GB | Fast |
| Shap-E | Great | FP16 | 4 GB | Fast |
| InstantMesh | Great | FP16 | 6 GB | Fast |
| Zero123++ v1.2 | Great | FP16 | 6 GB | Fast |
| LGM | Great | FP16 | 8 GB | Fast |
| One-2-3-45 | Great | FP16 | 8 GB | Fast |
| Hunyuan3D 2.1 | Good | FP16 | 10 GB | Fast |
| SV3D | Good | FP16 | 10 GB | Fast |
| Pixal3D | Good | FP8 | 10 GB | Moderate |
| Hunyuan3D 2.0 | Good | FP16 | 12 GB | Fast |
| TRELLIS.2 4B | Usable | FP8 | 12 GB | Moderate |
| SAM 3D Objects | Can't Run | — | needs 32 GB | — |
Audio
| Model | Rating | Runs as | VRAM used | Speed |
|---|---|---|---|---|
| MusicGen Small (300M) | Great | FP16 | 2 GB | Fast |
| XTTS v2 | Great | FP16 | 3 GB | Fast |
| Stable Audio 3 Small-SFX | Great | FP16 | 3 GB | Fast |
| Whisper Large v3 | Great | FP16 | 4 GB | Fast |
| Chatterbox Multilingual | Great | FP16 | 4 GB | Fast |
| MMAudio Large 44k | Great | FP16 | 4 GB | Fast |
| Stable Audio Open | Great | FP16 | 4 GB | Fast |
| Qwen3-TTS 1.7B | Great | FP16 | 5 GB | Fast |
| MusicGen Medium (1.5B) | Great | FP16 | 6 GB | Fast |
| IndexTTS 2.5 | Great | FP16 | 6 GB | Fast |
| VibeVoice 1.5B | Great | FP16 | 7 GB | Fast |
| Stable Audio 3 Medium | Great | FP16 | 7 GB | Fast |
| ACE-Step 1.5 Turbo | Great | FP16 | 8 GB | Fast |
| Bark | Great | FP16 | 8 GB | Fast |
| Fish Audio S2 Pro 4B | Good | GGUF Q4 | 7 GB | Moderate |
| MusicGen Large (3.3B) | Good | FP16 | 12 GB | Fast |
| YuE2 3B | Usable | FP8 | 12 GB | Moderate |
| MiniMax Music 3 | Usable | FP8 | 12 GB | Moderate |
Frequently asked questions
How much VRAM does the NVIDIA RTX 4070 Ti have?
The NVIDIA RTX 4070 Ti has 12 GB of VRAM with 504 GB/s of memory bandwidth.
What AI models can the NVIDIA RTX 4070 Ti run?
It can run 165 of the 168 AI models we track: 13 perfectly (S), 64 great (A), 68 good (B), 20 usable (C) and 0 barely (D). 3 need more VRAM than it has.
Can the NVIDIA RTX 4070 Ti run MiniMax H3 FL2VA?
Yes, in GGUF Q4 (rating C, usable).
Can the NVIDIA RTX 4070 Ti run MiniMax H3 Ref2VA?
Yes, in GGUF Q4 (rating C, usable).
Can the NVIDIA RTX 4070 Ti run Qwen-Image 2.1?
Yes, in FP8 (rating B, good).
Can the NVIDIA RTX 4070 Ti run Z-Image Turbo?
Yes, in FP16 (rating B, good).