CMP 170HX vs RTX 5070

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 50705 verified runs · best 185.8 tok/s

Specs and evidence

	CMP 170HX	RTX 5070
VRAM	64 GB	12 GB
Sourced runs on the board	16	5
Models measured	11	5
Best output tok/s	255.9 tok/s	185.8 tok/s
Highest measured peak VRAM	58.7 GB	9.0 GB
Street price (latest snapshot)	—	—

Sourced rows only. Peak VRAM is recorded by the person who ran the benchmark, so it is a floor for what the card can hold, not a specification. Price snapshots.

Head-to-head

On the 3 models measured on both cards, CMP 170HX leads on 1, RTX 5070 on 2. The widest gap is on Qwen3.8-27B: CMP 170HX 255.9 tok/s vs RTX 5070 48.3 tok/s (CMP 170HX 430% faster).

The measured matrix

Model	CMP 170HX	RTX 5070
Qwen3.8-27B	255.9 W4A16	48.3 UD-Q2_K_XL
NVIDIA-Nemotron-3-Nano-4B	—	185.8 Q4_K_M
Apodex-1.1-mini	—	182.3 IQ1_M
NVIDIA-Nemotron-3.5-Lightning-30B-A3B	165.1 Q8_0	—
Ornith-1.5-9B	61.5 BF16	102.0 Q4_K_M
Qwen3.6-35B-A3B	87.2 UD-Q8_K_XL	—
Ornith-1.5-35B-A3B	87.1 Q8_0	—
Qwen3.8-Flash-Next	46.1 UD-IQ1_S	81.8 Q2_0
Qwen3.8-27B-MTP	46.4 INT8-W8A16	—
Llama-4-Scout-17B-16E-Instruct	44.5 UD-Q3_K_XL	—
gemma-4-31B-it	28.3 NVFP4	—
granite-4.2-30b	23.7 Q8_0	—
Llama-3.3-70B-Instruct	20.1 NVFP4	—

Head-to-head (models tested on both): CMP 170HX 1 · RTX 5070 2 · tied 0 · data as of Oct 30, 2026. Full rules: methodology.

Related reading: RTX 5090 vs 3090 · quantization · Value index · Full board