1× RTX 30708 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-3070
Full price
€0.07
per hour, whole instance, 1 GPU
NVIDIA · Ampere
Rent RTX 3070
NVIDIA GeForce RTX 3070, 8 GB.inference smallRendering
Full price from
€0.07
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)
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 8 GB. That suits running a separate job on each card. Splitting one model across them works, but slowly.
Specifications
Consumer card: memory bandwidth and FLOPS figures come from NVIDIA's Ampere whitepaper (Table 10), not a product datasheet. No NVLink/SLI. FP16, BF16 and INT8 are with sparsity (FP16 with FP16 accumulate, BF16 with FP32 accumulate), following the RTX 3080 Ti entry. NVIDIA's Blackwell whitepaper gives slightly different RTX 3070 tensor figures (BF16 81.2, INT8 325); the Ampere whitepaper is used. fp64Tflops stays null because NVIDIA states the 1/64th FP64 ratio only for GA102, and this card uses GA104.
Memory
- VRAM
- 8 GB
- Memory type
- GDDR6
- Memory bandwidthHow fast the GPU reaches its memory. Often limits LLM speed
- 448 GB/s
- Bus width
- 256 bit
Compute
- FP32Speed at 32-bit precision
- 20.3 TFLOPS
- FP16Speed at 16-bit precision
- 162.6 TFLOPS
- BF1616-bit AI format. Standard for training and full-quality inference
- 81.3 TFLOPS
- INT88-bit integer speed, used by quantized models
- 325.2 TOPS
- FP8 support
- No
- FP4 support
- No
- Transformer EngineHardware that speeds up FP8 and FP4
- No
- Compute cores
- 5,888
Multi-GPU and connectivity
- NVSwitchLets every GPU in a server talk to every other at full speed
- No
- PCIe generation
- 4
Sharing and security
- MIG partitioningSplits one GPU into several isolated smaller GPUs
- No
- Max MIG instances
- 0
- Confidential computing
- No
Hardware
- Architecture
- Ampere
- Launch year
- 2020
- Process node
- Samsung 8N
- Transistors
- 17.4 billion
- Die size
- 392.5 mm²
- Power (TDP)Maximum power draw
- 220 W
- Cooling
- active
Video
- NVENC engines
- 1
- NVDEC engines
- 1
- AV1 encode
- No
- AV1 decode
- Yes
Sources (37)
- https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/rtx-3070-3070ti/
- https://nvidianews.nvidia.com/news/nvidia-delivers-greatest-ever-generational-leap-in-performance-with-geforce-rtx-30-series-gpus
- https://docs.nvidia.com/datacenter/tesla/mig-user-guide/supported-gpus.html
- https://www.nvidia.com/en-us/data-center/solutions/confidential-computing/
- https://docs.nvidia.com/590trd1-trusted-computing-solutions-release-notes.pdf
- https://www.nvidia.com/en-us/geforce/graphics-cards/compare/
- https://developer.nvidia.com/video-encode-decode-support-matrix
- https://www.nvidia.com/content/PDF/nvidia-ampere-ga-102-gpu-architecture-whitepaper-v2.pdf
- https://developer.nvidia.com/blog/nvidia-hopper-architecture-in-depth/
- https://developer.nvidia.com/blog/inside-nvidia-blackwell-ultra-the-chip-powering-the-ai-factory-era/