mDeBERTa-v3-base-mnli-xnli
by MoritzLaurer
279M params · zero-shot-classification · 296 likes · 144.4k downloads
mDeBERTa-v3-base-mnli-xnli is a 279M parameter model. At Q4 quantization it requires 0GB of VRAM. It runs comfortably on GeForce RTX 4090 (4709 tok/s), GeForce RTX 5090 (7062 tok/s), M4 Max 128GB (1723 tok/s).
Inference providers
| Provider | $/1M in | $/1M out | Throughput |
|---|---|---|---|
| HF Inference |
GPU compatibility
| GPU | VRAM | Q4 Decode | Verdict |
|---|---|---|---|
| GeForce RTX 4090 | 24GB | 4709 tok/s | comfortable |
| GeForce RTX 5090 | 32GB | 7062 tok/s | comfortable |
| M4 Max 128GB | 128GB | 1723 tok/s | comfortable |
| M4 Pro 48GB | 48GB | 861 tok/s | comfortable |
| M4 Pro 24GB | 24GB | 861 tok/s | comfortable |
| A100 PCIe 80 GB | 80GB | 8628 tok/s | comfortable |
| H100 SXM5 80 GB | 80GB | 17353 tok/s | comfortable |
| GeForce RTX 3090 | 24GB | 4163 tok/s | comfortable |
| Radeon RX 7900 XTX | 24GB | 3443 tok/s | comfortable |
| GeForce RTX 4080 | 16GB | 3342 tok/s | comfortable |