NVIDIA · Ampere

Rent RTX 3060

NVIDIA GeForce RTX 3060, 12 GB.inference smallRendering

Full price from

€0.04

in stock, per GPU-hour, as 1×

Spot from

—

in stock, per GPU-hour

Offers in stock

1 of 1

full price or spot

Providers in stock

1 of 1

full price or spot

Offers (1)

Cheapest in stock

NosanaIn stock · 7 freeStart right away

1× RTX 30608 GB VRAM

Nosana network, hosts worldwide

Details

Stock and price checked 8 min ago.

Self-service: you start it yourself, with no order or sales contact.

nvidia-3060

Full price

€0.04

per hour, whole instance, 1 GPU

View offer

Price and availability history

Price per GPU-hour

Excluding VAT.

Providers with stock

Each provider counts once. Full price and spot are separate.

Share of offers in stock

Full price. Offers we can't say are really sold there are left out until they've been in stock once.

Units free

Only providers that report a number. Each line is one provider and one server size, because their counts cover different things and can't be added up. Providers that only say yes or no are under Stock by provider.

Stock by provider

In stock, yes or no, for every provider selling this GPU. A filled band means in stock: green at full price, orange as spot.

NosanaIn stock

Reports a number

In stock as

The server sizes you can rent right now.

Nosana: 1×

Using several together

Can't be linked. This GPU has no NVLink, so several in one server only talk over PCIe. Each card keeps its own 12 GB. That suits running a separate job on each card. Splitting one model across them works, but slowly.

Specifications

Consumer card. Headline values are the 12 GB model; the 8 GB model has a 128-bit bus and 240 GB/s per TechPowerUp. NVIDIA publishes no FLOPS figures for this card. fp32Tflops is derived: 3584 CUDA cores x 2 FP32 operations per clock x 1.78 GHz boost = 12.8, a formula that reproduces NVIDIA's published FP32 figures for the RTX 3070, 3080 and 3090. FP16, BF16 and INT8 stay null because NVIDIA does not publish them for this card. fp64Tflops stays null because NVIDIA states the 1/64th FP64 ratio only for GA102. rtCores is derived: 3584 cores / 128 per SM = 28 SMs, one RT Core per SM as in NVIDIA's GA10x tables. memoryDataRateGbps is from TechPowerUp articles, not its GPU database page, which could not be opened; memoryBandwidthGBs is derived (192 bits x 15 Gbps / 8). pcieLanes is from a TechPowerUp review. processNode is NVIDIA's GA10x process, with TechPowerUp giving 8 nm for GA106. transistorsBillion and dieSizeMm2 stay null: TechPowerUp pages disagree on the transistor count (13.25 vs 12 billion) and no die size could be confirmed. cooling stays null because NVIDIA lists no Founders Edition for the 3060 and partner coolers vary. NVLink not supported.

Memory

VRAM: 12 GB
Memory type: GDDR6
Memory bandwidth How fast the GPU reaches its memory. Often limits LLM speed: 360 GB/s
Bus width: 192 bit

Compute

FP32 Speed at 32-bit precision: 12.8 TFLOPS
FP8 support: No
FP4 support: No
Transformer Engine Hardware that speeds up FP8 and FP4: No
Compute cores: 3,584

Multi-GPU and connectivity

NVSwitch Lets every GPU in a server talk to every other at full speed: No
PCIe generation: 4
PCIe lanes: 16

Sharing and security

MIG partitioning Splits one GPU into several isolated smaller GPUs: No
Max MIG instances: 0
Confidential computing: No

Hardware

Architecture: Ampere
Launch year: 2021
Process node: Samsung 8N
Power (TDP) Maximum power draw: 170 W

Video

NVENC engines: 1
NVDEC engines: 1
AV1 encode: No
AV1 decode: Yes

Sources (36)