This page answers nineninesix/kani-tts-2-en q4_k_m quantization queries with explicit calculations from our model requirement dataset and compatibility speed table.
Q4_K_M mapped to the same VRAM envelope as Q4 in current dataset.
Throughput data below uses available compatibility measurements/estimates and is sorted by tokens per second for this model.
Need general guidance? Review full methodology.
| GPU | VRAM | Quantization | Speed | Compatibility |
|---|---|---|---|---|
| AMD Instinct MI300X | 192GB | Q4 | 975 tok/s | View full compatibility |
| NVIDIA H200 SXM 141GB | 141GB | Q4 | 765 tok/s | View full compatibility |
| NVIDIA H100 SXM5 80GB | 80GB | Q4 | 618 tok/s | View full compatibility |
| AMD Instinct MI250X | 128GB | Q4 | 586 tok/s | View full compatibility |
| NVIDIA H100 PCIe 80GB | 80GB | Q4 | 385 tok/s | View full compatibility |
| NVIDIA A100 80GB SXM4 | 80GB | Q4 | 335 tok/s | View full compatibility |
| RTX 5090 | 32GB | Q4 | 333 tok/s | View full compatibility |
| AMD Instinct MI210 | 64GB | Q4 | 297 tok/s | View full compatibility |
| NVIDIA A100 40GB PCIe | 40GB | Q4 | 263 tok/s | View full compatibility |
| NVIDIA RTX 6000 Ada | 48GB | Q4 | 204 tok/s | View full compatibility |
| RTX 4090 | 24GB | Q4 | 199 tok/s | View full compatibility |
| AMD Radeon Pro W7900 | 48GB | Q4 | 189 tok/s | View full compatibility |