Yi-34B
by 01-ai
34.4B params · text-generation · 1.3k likes · 10.9k downloads
Yi-34B is a 34.4B parameter model. At Q4 quantization it requires 17GB of VRAM. It runs comfortably on GeForce RTX 4090 (38 tok/s), GeForce RTX 5090 (57 tok/s), A100 PCIe 80 GB (69 tok/s).
GPU compatibility
| GPU | VRAM | Q4 Decode | Verdict |
|---|---|---|---|
| GeForce RTX 4090 | 24GB | 38 tok/s | comfortable |
| GeForce RTX 5090 | 32GB | 57 tok/s | comfortable |
| M4 Max 128GB | 128GB | 13 tok/s | tight |
| M4 Pro 48GB | 48GB | 6 tok/s | tight |
| M4 Pro 24GB | 24GB | 6 tok/s | tight |
| A100 PCIe 80 GB | 80GB | 69 tok/s | comfortable |
| H100 SXM5 80 GB | 80GB | 140 tok/s | comfortable |
| GeForce RTX 3090 | 24GB | 33 tok/s | comfortable |
| Radeon RX 7900 XTX | 24GB | 27 tok/s | tight |
| M4 Max 64GB | 64GB | 13 tok/s | tight |