snowflake-arctic-embed-l
by Snowflake
334M params · sentence-similarity · 100 likes · 22.4k downloads
snowflake-arctic-embed-l is a 334M parameter model. At Q4 quantization it requires 0GB of VRAM. It runs comfortably on GeForce RTX 4090 (3930 tok/s), GeForce RTX 5090 (5893 tok/s), M4 Max 128GB (1438 tok/s).
Inference providers
| Provider | $/1M in | $/1M out | Throughput |
|---|---|---|---|
| HF Inference |
GPU compatibility
| GPU | VRAM | Q4 Decode | Verdict |
|---|---|---|---|
| GeForce RTX 4090 | 24GB | 3930 tok/s | comfortable |
| GeForce RTX 5090 | 32GB | 5893 tok/s | comfortable |
| M4 Max 128GB | 128GB | 1438 tok/s | comfortable |
| M4 Pro 48GB | 48GB | 719 tok/s | comfortable |
| M4 Pro 24GB | 24GB | 719 tok/s | comfortable |
| A100 PCIe 80 GB | 80GB | 7200 tok/s | comfortable |
| H100 SXM5 80 GB | 80GB | 14482 tok/s | comfortable |
| GeForce RTX 3090 | 24GB | 3474 tok/s | comfortable |
| Radeon RX 7900 XTX | 24GB | 2873 tok/s | comfortable |
| GeForce RTX 4080 | 16GB | 2789 tok/s | comfortable |