Confirm condition, inspection evidence and commercial terms in the quote.
🖥️

NVIDIA

Photo Coming Soon

GPUs & Accelerators

NVIDIA

NVIDIA V100 32GB — Used & Refurbished (NVIDIA)

Part No. NVIDIA V100 32GB

Price on request

Ask for a quote tied to the required configuration and condition.

Contact for availability. Stock and condition require confirmation for a specific unit.

Place a Bid or Ask

Talk to a specialist · parts@caladansemi.com

Get a quote for this partPricing and availability to be confirmed

✓Request unit-specific inspection and test records
✓Confirm included documentation and accessories
✓Agree delivery, payment and return terms in the quote

How Much Does NVIDIA V100 32GB Cost in 2026?

In 2026, used NVIDIA V100 32GB GPUs typically range from $3,000 to $25,000, depending on condition and provenance. Refurbished units with warranty (e.g., from cloud vendors like AWS or Azure) average $12,000–$18,000, while vendor-certified (e.g., NVIDIA-authorized reconditioned) models hit $20,000–$25,000. Raw used parts (no testing) drop to $3,000–$6,000, but failure rates exceed 20% in untested batches. Compared to the 2023 OEM list price of $9,000–$12,000, buyers save 10–60% depending on the grade. Caveat: Power supply upgrades (minimum 800W 80+ Gold) are often required, adding $200–$500 to the total cost.

NVIDIA V100 32GB Specifications and Compatible Systems

The V100 32GB features 32GB HBM2 memory, 70 RT cores, 640 Tensor cores, and 5,120 CUDA cores, with a peak performance of 14 TeraFLOPS (FP16). It requires a PCIe 3.0 x16 slot and dual 8-pin power connectors. Compatible systems include Dell PowerEdge R750xd (2021), HPE ProLiant DL380 Gen10 (2019), and Lenovo SR670 (2020). Avoid systems with PCIe 4.0 unless the motherboard’s BIOS is patched for backward compatibility. Note: The V100’s high TDP (250–300W) demands liquid cooling in dense racks; air-cooled variants risk thermal throttling above 85°C.

What to Check Before Buying a Used NVIDIA V100 32GB in 2026

  1. Physical Inspection: Look for bulging capacitors on the VRM (voltage regulator module) and HBM2 heatsink warping.
  2. VRAM Test: Run MemTestG80 for 24 hours; error rates above 0.01% indicate failing HBM2.
  3. Thermal Performance: Use nvidia-smi to log temperatures under CUDA stress. Sustained >85°C without liquid cooling is a red flag.
  4. Fan and Power Delivery: Check fan RPM via NVML (minimum 1,200 RPM under load). Verify power supply delivers 50A+ on the 12V rail.
  5. **Firmware

Quick Info

ConditionConfirm in quote
CategoryGPUs & Accelerators
Secondary Market

Place a Bid or Ask

Submit your buying or selling requirements for NVIDIA V100 32GB — Used & Refurbished (NVIDIA).

Want to Buy
Want to Sell

Confirm export requirements and payment terms before any transaction · parts@caladansemi.com