The honest, $0-markup cost to run DeepSeek-R1 (DeepSeek, 671B) and DeepSeek R1 0528 (deepseek-ai, 684.5B), side by side — free on your own machine, on your own rented GPU at the vendor's price, or via your own API key.
| DeepSeek-R1 | DeepSeek R1 0528 | |
|---|---|---|
| Parameters | 671B | 684.5B |
| Context window | 163.8K | 163.8K |
| License | Commercial OK | Commercial OK |
| VRAM to run | ~453 GB (Q4_K_M) | ~489 GB (Q4_K_M) |
| Rent a GPU | $9.73/hr | $9.73/hr |
| Your API key | $1.60/1M | $1.33/1M |
VRAM, size, context and license are facts from the catalog and the shared cost engine; API prices are real where we have them (labeled "est." otherwise). We don't rank model quality or quote benchmarks here.
Full cost breakdown: DeepSeek-R1 →
Full cost breakdown: DeepSeek R1 0528 →
Open the free Spanvero advisor → to compare them live for your exact workload and hardware.
Spanvero · All comparisons · Prices as of 2026-06-17. $0 markup, your own accounts, we never resell compute. © 2026 Cynosure LLC.