Gemma-SEA-LION-v4-27B-IT
by aisingapore
27.4B params · text-generation · 21 likes · 1.3k downloads
Gemma-SEA-LION-v4-27B-IT is a 27.4B parameter model. At Q4 quantization it requires 14GB of VRAM. It runs comfortably on GeForce RTX 4090 (47 tok/s), GeForce RTX 5090 (71 tok/s), A100 PCIe 80 GB (87 tok/s).
Inference providers
| Provider | $/1M in | $/1M out | Throughput |
|---|---|---|---|
| Featherless | |||
| Public AI | 52 tok/s |
GPU compatibility
| GPU | VRAM | Q4 Decode | Verdict |
|---|---|---|---|
| GeForce RTX 4090 | 24GB | 47 tok/s | comfortable |
| GeForce RTX 5090 | 32GB | 71 tok/s | comfortable |
| M4 Max 128GB | 128GB | 17 tok/s | tight |
| M4 Pro 48GB | 48GB | 8 tok/s | tight |
| M4 Pro 24GB | 24GB | 8 tok/s | tight |
| A100 PCIe 80 GB | 80GB | 87 tok/s | comfortable |
| H100 SXM5 80 GB | 80GB | 176 tok/s | comfortable |
| GeForce RTX 3090 | 24GB | 42 tok/s | comfortable |
| Radeon RX 7900 XTX | 24GB | 34 tok/s | comfortable |
| M4 Max 64GB | 64GB | 17 tok/s | tight |