NVIDIA GeForce RTX 5090 vs NVIDIA GeForce RTX 5060 Ti 16GB
For running models locally, priced in Canada. At 64k context the first runs 27 models fully in memory, the second 10.
Specifications
| Spec | NVIDIA GeForce RTX 5090 | NVIDIA GeForce RTX 5060 Ti 16GB |
|---|---|---|
| GPU | GeForce RTX 5090 | GeForce RTX 5060 Ti 16GB |
| Memory | 32 GB GDDR7 | 16 GB GDDR7 |
| Memory bus | 512-bit | 128-bit |
| Memory bandwidth | 1,792 GB/s | 448 GB/s |
| FP16 compute | 209 TFLOPS | 47 TFLOPS |
| Software | CUDA | CUDA |
| Board power | 575 W | 180 W |
| PCIe generation | Gen 5 | Gen 5 |
| PCIe lanes | x16 | x8 |
What each runs
Estimated decode speed with the most precise quant that fits each one est.
| Model | NVIDIA GeForce RTX 5090 | NVIDIA GeForce RTX 5060 Ti 16GB |
|---|---|---|
| Qwen 3.8 27B | 42tok/sUD-Q6_K | offload |
| Qwen 3.6 27B | 41tok/sQ6_K | offload |
| Gemma 4 31B | 35tok/sQ6_K | offload |
| Qwen 3.6 35B-A3B | 173tok/sUD-Q6_K | offload |
| Qwen 3.5 27B | 41tok/sQ6_K | offload |
| Gemma 4 26B-A4B | 117tok/sQ8_0 | offload |
| Qwen 3.5 35B-A3B | 175tok/sQ6_K | offload |
| GLM 4.7 Flash | 87tok/sQ6_K | offload |
| Devstral Small 2 24B | 37tok/sQ6_K | offload |
| Kimi Linear 48B-A3B | 277tok/sQ4_K_M | offload |
| gpt-oss 20B | 165tok/sMXFP4 | 41tok/sMXFP4 |
| Qwen3 Coder 30B-A3B | 75tok/sQ6_K | offload |
| Gemma 4 12B | 79tok/sQ8_0 | 20tok/sQ8_0 |
| Qwen 3.5 9B | 95tok/sQ8_0 | 24tok/sQ8_0 |
| Gemma 3 27B | 40tok/sQ6_K | offload |
| Mistral Small 3.2 24B | 37tok/sQ6_K | offload |
| Ministral 3 14B | 44tok/sQ8_0 | offload |
| Qwen3 14B | 42tok/sQ8_0 | offload |
| DeepSeek R1 0528 Qwen3 8B | 61tok/sQ8_0 | 19tok/sQ4_K_M |
| Qwen3 8B | 61tok/sQ8_0 | 19tok/sQ4_K_M |
| Phi-4 14B | 38tok/sQ8_0 | offload |
| Qwen 3.5 4B | 166tok/sQ8_0 | 42tok/sQ8_0 |
| Ministral 3 8B | 61tok/sQ8_0 | 18tok/sQ5_K_M |
| Granite 4.2 8B | 55tok/sQ8_0 | offload |
| Gemma 4 E4B | 110tok/sQ8_0 | 28tok/sQ8_0 |
| LFM2.5 8B-A1B | 156tok/sQ8_0 | 39tok/sQ8_0 |
| Llama 3.1 8B | 73tok/sQ6_K | 18tok/sQ6_K |
| Qwen 3.5 122B-A10B | offload | won't run |
| Qwen3 Coder Next | offload | won't run |
| Nemotron 3 Super 120B-A12B | offload | won't run |
| Mistral Small 4 119B | offload | offload |
| GLM 4.5 Air | offload | won't run |
| Qwen3 Next 80B-A3B | offload | won't run |
| Nemotron 3.5 Lightning 30B-A3B | offload | offload |
| Qwen3 32B | offload | offload |
| DeepSeek R1 Distill Qwen 32B | offload | offload |
| DeepSeek V4 Flash | won't run | won't run |
| GLM 5.3 Flash | won't run | won't run |
| Qwen 3.8 Flash Next | won't run | won't run |
| MiniMax M2.7 | won't run | won't run |
| gpt-oss 120B | won't run | won't run |
| Llama 3.3 70B | won't run | won't run |