CMP 170HX vs RTX 3090

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 309018 verified runs · best 709.5 tok/s

Specs and evidence

	CMP 170HX	RTX 3090
VRAM	64 GB	24 GB
Sourced runs on the board	16	18
Models measured	11	11
Best output tok/s	255.9 tok/s	709.5 tok/s
Highest measured peak VRAM	58.7 GB	22.0 GB
Street price (latest snapshot)	—	$1,720 used

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 2, RTX 3090 on 1. The widest gap is on NVIDIA-Nemotron-3.5-Lightning-30B-A3B: CMP 170HX 165.1 tok/s vs RTX 3090 353.7 tok/s (RTX 3090 114% faster).

The measured matrix

Model	CMP 170HX	RTX 3090
LFM2.5-1.2B-Instruct	—	709.5 Q4_K_M
Qwen3.5-0.8B-MTP	—	476.1 Q6_K_XL
NVIDIA-Nemotron-3.5-Lightning-30B-A3B	165.1 Q8_0	353.7 NVFP4
LFM2.5-2.6B	—	305.6 Q4_K_M
Qwen3.8-27B	255.9 W4A16	164.6 EXL3-3.5bpw
Qwen3.5-4B-MTP	—	240.4 Q6_K_XL
Qwen3.5-9B-MTP	—	164.3 Q6_K_XL
Qwen3.6-35B-A3B	87.2 UD-Q8_K_XL	—
Ornith-1.5-35B-A3B	87.1 Q8_0	—
Ornith-1.5-9B	61.5 BF16	—
Qwen3.8-27B-MTP	46.4 INT8-W8A16	—
Qwen3.8-Flash-Next	46.1 UD-IQ1_S	38.6 EXL3 2.50bpw
Llama-4-Scout-17B-16E-Instruct	44.5 UD-Q3_K_XL	—
Ornstein3.6-27B-MTP-NSC-ACE-SABER	—	32.9 Q4_K_M
gemma-4-31B-it	28.3 NVFP4	—
GLM-5.3-Flash	—	28.1 EXL3 3.05 bpw
DeepSeek-V4.1-Flash	—	26.0 EXL3 3.0 bpw
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 2 · RTX 3090 1 · tied 0 · data as of Oct 17, 2026. Full rules: methodology.

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