DiggTop ⌕

CMP 170HX x2 vs CMP 170HX x4

Pick a different pair →

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.

Sourced data on this pair is thin — most of the board is still partial. Submit your own run and help fill the matrix: one prompt, three steps, no downloads.
CMP 170HX x25 sourced runs · best 186.9 tok/s
CMP 170HX x42 verified runs · best 264.6 tok/s

Specs and evidence

CMP 170HX x2CMP 170HX x4
VRAM64 GB64 GB
Sourced runs on the board52
Models measured52
Best output tok/s186.9 tok/s264.6 tok/s
Highest measured peak VRAM118.0 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.

The measured matrix

ModelCMP 170HX x2CMP 170HX x4
GLM-5.3-Flash—264.6 W4A16
Qwen3.8-Flash-Next186.9 W4A16-FP8PLE—
DeepSeek-V4-Flash-0731—92.1 FP8
Laguna-S-2.163.3 Q4_K_M—
Hy-MT2-30B-A3B61.4 FP8—
Hy327.6 Q2_K_XL—
Baichuan-M3-235B26.1 Q3_K_M—

7 rows · 5 models on CMP 170HX x2, 2 on CMP 170HX x4 — no verified overlap yet, each side shows its own bests · data as of Oct 17, 2026. Full rules: methodology.

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

↑