distilroberta-finetuned-financial-news-sentiment-analysis
by mrm8488
82M params · text-classification · 445 likes · 409.7k downloads
distilroberta-finetuned-financial-news-sentiment-analysis is a 82M parameter model. At Q4 quantization it requires 0GB of VRAM. It runs comfortably on GeForce RTX 4090 (15988 tok/s), GeForce RTX 5090 (23976 tok/s), M4 Max 128GB (5850 tok/s).
Inference providers
| Provider | $/1M in | $/1M out | Throughput |
|---|---|---|---|
| HF Inference |
GPU compatibility
| GPU | VRAM | Q4 Decode | Verdict |
|---|---|---|---|
| GeForce RTX 4090 | 24GB | 15988 tok/s | comfortable |
| GeForce RTX 5090 | 32GB | 23976 tok/s | comfortable |
| M4 Max 128GB | 128GB | 5850 tok/s | comfortable |
| M4 Pro 48GB | 48GB | 2925 tok/s | comfortable |
| M4 Pro 24GB | 24GB | 2925 tok/s | comfortable |
| A100 PCIe 80 GB | 80GB | 29293 tok/s | comfortable |
| H100 SXM5 80 GB | 80GB | 58917 tok/s | comfortable |
| GeForce RTX 3090 | 24GB | 14136 tok/s | comfortable |
| Radeon RX 7900 XTX | 24GB | 11690 tok/s | comfortable |
| GeForce RTX 4080 | 16GB | 11347 tok/s | comfortable |