CMP 170HX vs RTX 5070 12GB + RTX 5060 Ti 16GB x2
Best sourced output tok/s per model (single-stream generation) — engine and version, quantization, context and a source link on every row. Hover a value for the quant.
CMP 170HX16 sourced runs · best 255.9 tok/s
RTX 5070 12GB + RTX 5060 Ti 16GB x21 verified run · best 66.0 tok/s
Specs and evidence
| CMP 170HX | RTX 5070 12GB + RTX 5060 Ti 16GB x2 | |
|---|---|---|
| VRAM | 64 GB | 12 GB |
| Sourced runs on the board | 16 | 1 |
| Models measured | 11 | 1 |
| Best output tok/s | 255.9 tok/s | 66.0 tok/s |
| Highest measured peak VRAM | 58.7 GB | — |
| Street price (latest snapshot) | — | — |
Head-to-head
Only one model has been measured on both cards (Qwen3.8-Flash-Next). On that single data point CMP 170HX records 46.1 tok/s and RTX 5070 12GB + RTX 5060 Ti 16GB x2 records 66.0 tok/s — treat it as one data point, not a ranking.
The measured matrix
| Model | CMP 170HX | RTX 5070 12GB + RTX 5060 Ti 16GB x2 |
|---|---|---|
| Qwen3.8-27B | 255.9 | — |
| NVIDIA-Nemotron-3.5-Lightning-30B-A3B | 165.1 | — |
| Qwen3.6-35B-A3B | 87.2 | — |
| Ornith-1.5-35B-A3B | 87.1 | — |
| Qwen3.8-Flash-Next | 46.1 | 66.0 |
| Ornith-1.5-9B | 61.5 | — |
| Qwen3.8-27B-MTP | 46.4 | — |
| Llama-4-Scout-17B-16E-Instruct | 44.5 | — |
| gemma-4-31B-it | 28.3 | — |
| granite-4.2-30b | 23.7 | — |
| Llama-3.3-70B-Instruct | 20.1 | — |