§01·spec · /gpus
RTX 5070 Ti
nvidia50 series16GB VRAM
46 open-weights AI models run on the RTX 5070 Ti — see which fit, how fast they go, and the VRAM each needs.
§02·models that run on this GPU
46 run · 47 testedLLM · 7
| Model | Best speed | Min VRAM | Works | Evidence | |
|---|---|---|---|---|---|
| gpt-oss 20B | 156tokens/s | 16GB | ✓ | 2benchesrecipe | check ↗ |
| Qwen3-8B | 120.5tokens/s | 16GB | ✓ | 2benchesrecipe | check ↗ |
| Qwen3 14B | 74.3tokens/s | 16GB | ✓ | 2benchesrecipe | check ↗ |
| Apodex 1.1 mini | 16GB | ✓ | recipe | check ↗ | |
| Llama 3.1 8B | 10GB | ✓ | recipe | check ↗ | |
| Nanbeige4.2 3B | 16GB | ✓ | recipe | check ↗ | |
| Ornith 1.0 9B | 12GB | ✓ | recipe | check ↗ |
- Best speed
- 156tokens/s
- Min VRAM
- 16GB
- Evidence
- 2benches
- Best speed
- 120.5tokens/s
- Min VRAM
- 16GB
- Evidence
- 2benches
- Best speed
- 74.3tokens/s
- Min VRAM
- 16GB
- Evidence
- 2benches
- Best speed
- Min VRAM
- 16GB
- Best speed
- Min VRAM
- 10GB
- Best speed
- Min VRAM
- 16GB
- Best speed
- Min VRAM
- 12GB
Multimodal · 7
| Model | Best speed | Min VRAM | Works | Evidence | |
|---|---|---|---|---|---|
| Fara1.5-4B | 16GB | ✓ | recipe | check ↗ | |
| Fara1.5-9B | 16GB | ✓ | recipe | check ↗ | |
| Gemma 4 E4B-IT | 6GB | ✓ | recipe | check ↗ | |
| MiniMind-O | 4GB | ✓ | recipe | check ↗ | |
| MOSS-Audio | 12GB | ✓ | recipe | check ↗ | |
| Qwen3.8 27B | 16GB | ✓ | recipe | check ↗ | |
| Voxtral Mini 3B | 10GB | ✓ | recipe | check ↗ |
Image · 12
| Model | Best speed | Min VRAM | Works | Evidence | |
|---|---|---|---|---|---|
| Anima | 7GB | ✓ | recipe | check ↗ | |
| Chroma V48 | 14GB | ✓ | recipe | check ↗ | |
| ERNIE-Image-Turbo | 12GB | ✓ | recipe | check ↗ | |
| Flux.2-Klein-4B | 13GB | ✓ | recipe | check ↗ | |
| HiDream-O1-Image | 10GB | ✓ | recipe | check ↗ | |
| Juggernaut Z | 16GB | ✓ | recipe | check ↗ | |
| Krea 2 | 16GB | ✓ | recipe | check ↗ | |
| LongCat Image | 16GB | ✓ | recipe | check ↗ | |
| Qwen-Image | 13GB | ✓ | recipe | check ↗ | |
| Qwen-Image-2.1 | 12GB | ✓ | recipe | check ↗ | |
| SenseNova U1 | 12GB | ✓ | recipe | check ↗ | |
| Z-Image Turbo | 16GB | ✓ | recipe | check ↗ |
- Best speed
- Min VRAM
- 7GB
- Best speed
- Min VRAM
- 14GB
- Best speed
- Min VRAM
- 12GB
- Best speed
- Min VRAM
- 13GB
- Best speed
- Min VRAM
- 10GB
- Best speed
- Min VRAM
- 16GB
- Best speed
- Min VRAM
- 16GB
- Best speed
- Min VRAM
- 16GB
- Best speed
- Min VRAM
- 13GB
- Best speed
- Min VRAM
- 12GB
- Best speed
- Min VRAM
- 12GB
- Best speed
- Min VRAM
- 16GB
Video · 7
3D · 3
| Model | Best speed | Min VRAM | Works | Evidence | |
|---|---|---|---|---|---|
| Hunyuan3D | 10GB | ✓ | recipe | check ↗ | |
| TRELLIS image-large | 16GB | ✓ | recipe | check ↗ | |
| Waypoint 1.5 | 12GB | ✓ | recipe | check ↗ |
TTS · 6
Music · 3
| Model | Best speed | Min VRAM | Works | Evidence | |
|---|---|---|---|---|---|
| ACE-Step 1.5 XL | 12GB | ✓ | recipe | check ↗ | |
| Foundation-1 | 8GB | ✓ | recipe | check ↗ | |
| YuE2-3B | 16GB | ✓ | recipe | check ↗ |
§03·tested recipes
showing 6 of 46- imageintermediate12GB+recipe
Qwen-Image-2.1 on RTX 5070 Ti: int8 template, 16 GB budget and an edit-grid sweep by one user
- musicintermediate16GB+recipe
YuE2-3B on RTX 5070 Ti: same pins as the measured card, twice the bus
- llmadvanced16GB+recipe
Apodex 1.1 mini on RTX 5070 Ti: a 36B agent at 128K, and where DDR5 flips the bottleneck
- multimodaladvanced16GB+recipe
Qwen3.8-27B on RTX 5070 Ti: a vision-capable 27B inside 16 GB with llama.cpp
- videoadvanced16GB+recipe
LTX-2.5 on RTX 5070 Ti: 22B audio-video in 16 GB, and the decode trap three owners hit
- llmintermediate16GB+recipe
Nanbeige4.2-3B on RTX 5070 Ti: 96K context, and what twice the bandwidth cannot buy