self-hosted
/ai
GPUs
Models
Compare
Tools
Search
⌘K
§01
·
spec · /gpus
RTX 5070
nvidia
50 series
12GB VRAM
models (2)
§02
·
models that run on this GPU
2 total
Model
Vertical
Best speed
Min VRAM
Works
Benchmarks
Qwen3-8B
llm
85.8
tokens/s
12
GB
✓
2
check ↗
Qwen3 14B
llm
54.2
tokens/s
12
GB
✓
2
check ↗