Head to head
RTX 5080 vs RTX 4060 Ti 16GB
Both hold 16 GB, so capacity is a wash. The RTX 5080 has more bandwidth at 960 GB/s, which decides how fast tokens come out once a model fits.
| Model (Q4_K_M, 8K) | RTX 5080 | tok/s | RTX 4060 Ti 16GB | tok/s |
|---|---|---|---|---|
| Llama 3.1 8B | Runs comfortably | 121 | Runs comfortably | 33.6 |
| Llama 3.1 70B | Won't fit | 1.35 | Won't fit | 1.17 |
| Llama 3.2 3B | Runs comfortably | 266 | Runs comfortably | 74.8 |
| Llama 3.2 1B | Runs comfortably | 700 | Runs comfortably | 200 |
| Llama 4 Scout 109B-A17B | Won't fit | 4.91 | Won't fit | 4.28 |
| Qwen3 8B | Runs comfortably | 117 | Runs comfortably | 32.5 |
| Qwen3 14B | Runs comfortably | 68.7 | Runs comfortably | 18.9 |
| Qwen3 32B | Won't fit | 4.98 | Won't fit | 3.71 |
| Qwen3 4B | Runs comfortably | 212 | Runs comfortably | 59.6 |
| Qwen3 30B-A3B | Won't fit | 46.2 | Won't fit | 32.0 |
| Qwen3 235B-A22B | Won't fit | 3.36 | Won't fit | 3.01 |
| Qwen2.5-Coder 32B | Won't fit | 4.98 | Won't fit | 3.71 |
| Qwen2.5 7B | Runs comfortably | 135 | Runs comfortably | 37.3 |
| Qwen2.5 72B | Won't fit | 1.31 | Won't fit | 1.13 |
| Gemma 3 4B | Runs comfortably | 226 | Runs comfortably | 63.6 |
| Gemma 3 12B | Runs comfortably | 82.3 | Runs comfortably | 22.8 |
Specifications
| RTX 5080 | RTX 4060 Ti 16GB | |
|---|---|---|
| Memory | 16 GB | 16 GB |
| Bandwidth | 960 GB/s | 288 GB/s |
| FP16 compute | 112 TFLOPS | 44 TFLOPS |
| Launch price | $999 | $499 |
| Architecture | Blackwell | Ada |