CMP 170HX 40GB Price & Used Prices: Which Local LLMs Can It Run?
The CMP 170HX 40GB has 40GB of HBM2e memory with 1,560 GB/s of bandwidth. No price data yet. At Q4 quantization it runs models up to 32B parameters at roughly 54 tokens/s.
CMP 170HX 40GB price history
eBay
Full CMP 170HX 40GB specifications
What LLMs can the CMP 170HX 40GB run?
Versions released since 2026 with up to 50B parameters| Model | 2-bit | 4-bit | 8-bit | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Fits | Speed t/s | Context | Fits | Speed t/s | Context | Fits | Speed t/s | Context | |
| GemmaGemma 4 E4B | ✓ | 257.4 | 128K | ✓ | 189.5 | 128K | ✓ | 112.8 | 128K |
| QwenQwen3.5 9B | ✓ | 216.6 | 128K | ✓ | 158.9 | 128K | ✓ | 94.1 | 128K |
| GemmaGemma 4 12B | ✓ | 157.8 | 32K | ✓ | 118.8 | 32K | ✓ | 72.6 | 32K |
| GemmaGemma 4 26B A4B | ✓ | 160 | 32K | ✓ | 144 | 32K | ✓ | 114.4 | 32K |
| QwenQwen3.6 27B | ✓ | 87.9 | 32K | ✓ | 61.7 | 32K | ✓ | 34.9 | 32K |
| QwenQwen3.8 27B | ✓ | 87.9 | 32K | ✓ | 61.7 | 32K | ✓ | 34.9 | 32K |
| GLMGLM 4.7 Flash | ✓ | 191.1 | 128K | ✓ | 173.8 | 128K | ✓ | 140.8 | 128K |
| GemmaGemma 4 31B | ✓ | 66.2 | 8K | ✓ | 48.6 | 8K | ✗ | — | — |
| QwenQwen3.6 35B A3B | ✓ | 187.6 | 128K | ✓ | 170.9 | 128K | ✓ | 138.9 | 8K |
Fits when weights + KV cache + runtime overhead is at or below usable memory, single card. "Offload" on MoE models means it runs with expert weights in system RAM (only attention layers and KV cache stay in VRAM); that speed assumes 70 GB/s RAM bandwidth. The context column is the largest tier that fits.
Prices by platform
Buy links may carry affiliate tags; they do not change the price you pay.
FAQ
How much VRAM does the CMP 170HX 40GB have?
The CMP 170HX 40GB has 40GB of HBM2e memory with 1,560 GB/s of bandwidth.
Can the CMP 170HX 40GB run a 70B model?
Not on a single card. At Q4_K_M with an 8K context a 70B dense model needs about 43GB, more than the CMP 170HX 40GB's 40GB.
How much does a CMP 170HX 40GB cost?
No price data yet; it will appear here once the feed is live.