fineweb-edu-classifier
by HuggingFaceFW
109M params · text-classification · 208 likes · 10.0k downloads
fineweb-edu-classifier is a 109M parameter model. At Q4 quantization it requires 0GB of VRAM. It runs comfortably on GeForce RTX 4090 (11992 tok/s), GeForce RTX 5090 (17984 tok/s), M4 Max 128GB (4388 tok/s).
Inference providers
| Provider | $/1M in | $/1M out | Throughput |
|---|---|---|---|
| HF Inference |
GPU compatibility
| GPU | VRAM | Q4 Decode | Verdict |
|---|---|---|---|
| GeForce RTX 4090 | 24GB | 11992 tok/s | comfortable |
| GeForce RTX 5090 | 32GB | 17984 tok/s | comfortable |
| M4 Max 128GB | 128GB | 4388 tok/s | comfortable |
| M4 Pro 48GB | 48GB | 2194 tok/s | comfortable |
| M4 Pro 24GB | 24GB | 2194 tok/s | comfortable |
| A100 PCIe 80 GB | 80GB | 21972 tok/s | comfortable |
| H100 SXM5 80 GB | 80GB | 44193 tok/s | comfortable |
| GeForce RTX 3090 | 24GB | 10603 tok/s | comfortable |
| Radeon RX 7900 XTX | 24GB | 8768 tok/s | comfortable |
| GeForce RTX 4080 | 16GB | 8511 tok/s | comfortable |