Can the A800 PCIE run a 70B model?

Yes, quantized. 70B weights take roughly 35–40 GB at 4-bit, so they fit in 80 GB with headroom for context. Running 70B in 16-bit needs about 140 GB, which points to the H200 NVL.

How does the A800 PCIE differ from the A100 PCIE?

Same Ampere generation and 15.6 TFLOPS FP32, but the A800 PCIE is listed with 80 GB instead of 40 GB and 1598 GB/s instead of 1313 GB/s, so it fits larger models and serves them a little faster.

Should I choose the A800 PCIE or the H100 PCIE?

Both have 80 GB. The Hopper-based H100 PCIE adds 41 TFLOPS FP32 and 1760 GB/s, so it is faster per hour of work; the A800 PCIE makes sense when the live price is lower and Ampere speed is enough.

Which software can I run on the A800 PCIE?

The console offers Ubuntu 22.04 + CUDA and PyTorch + JupyterLab templates, and your own public Docker image at higher account levels. Tools such as vLLM, SGLang, Ollama, Unsloth or ComfyUI you install over SSH or include in your image.

GPU Cloud · NVIDIA · Ampere

Rent an NVIDIA A800 PCIE GPU with 80 GB VRAM

Rent A800 PCIE GPUs by the hour on Cloudj: 80 GB VRAM on NVIDIA Ampere with about 1598 GB/s memory bandwidth, so 70B-class LLMs fit on one card. It is aimed at AI teams that need double the A100 40 GB memory for serving, fine-tuning and long-context work, with 15.6 TFLOPS FP32 and a 300 W power limit. Pick a live configuration, start from the Ubuntu 22.04 + CUDA or PyTorch + JupyterLab template and connect over SSH with your key.

Rent A800 PCIE All GPUs

A800 PCIE specifications

VRAM per GPU

80 GB

Architecture

NVIDIA Ampere

Compute capability

FP32 compute

≈ 15.6 TFLOPS

Memory bandwidth

≈ 1598 GB/s

Power limit

300 W

Values reported by the infrastructure provider for machines currently listed; real performance varies by machine.

Live A800 PCIE configurations and prices

Out of stock right now

Browse available GPUs

What you can run on the A800 PCIE cloud GPU

70B models on a single GPU

Serve a 4-bit 70B model with room left for KV cache, or a 30B–34B model in 16-bit, using vLLM or SGLang.

Long-context inference

80 GB leaves space for long prompts and many concurrent sequences that would overflow a 40 GB card.

Fine-tuning mid-size LLMs

Run LoRA on 13B–34B models in BF16 or QLoRA on larger ones with Unsloth, without sharding across GPUs.

Memory-heavy data science

Keep big tensors, graphs or feature matrices resident in GPU memory for RAPIDS or PyTorch workloads.

Is the A800 PCIE the right GPU for your workload?

Choose the A800 PCIE when your primary requirement is 80 GB of memory capacity for large-scale AI models rather than raw single-precision compute. If your workload fits into a smaller memory footprint, a lower-VRAM workstation or consumer GPU is more appropriate, whereas compute-bound tasks may benefit from newer generation datacenter architectures.

A800 PCIE vs similar cloud GPUs

GPU	VRAM per GPU	FP32 compute	Power limit
NVIDIA A800 PCIE	80 GB	≈ 15.6 TFLOPS	300 W
NVIDIA H100 PCIE	80 GB	≈ 41 TFLOPS	350 W	Details →
NVIDIA H100 SXM	80 GB	≈ 53.5 TFLOPS	700 W	Details →
NVIDIA H100 NVL	94 GB	≈ 48.3 TFLOPS	400 W	Details →
NVIDIA CMP 170HX	64 GB	≈ 12.1 TFLOPS	250 W	Details →

Guides for A800 PCIE

GPU cloud Asia guide: Japan, Singapore, Korea and nearbyA practical GPU cloud Asia guide: which country codes to search for East and Southeast Asian users, a fallback order, and how to pair a GPU with a VPS.GPU cloud by location: how to pick a country for your GPUHow GPU cloud by location works on Cloudj: filter a live list of machines by country code, decide between near users, lowest price or data rules.GPU cloud Europe: choosing a country for your AI workloadsA practical guide to GPU cloud Europe searches: filter by several country codes at once, weigh Western, Nordic and Eastern hosts, and plan where data lives.GPU cloud North America: renting GPUs in the US and CanadaA GPU cloud North America guide: how to search US and CA hosts, read the results, pick a VPS front end and decide when a US GPU suits users elsewhere.

All articles →

A800 PCIE cloud GPU: frequently asked questions

Rent the A800 PCIE cloud GPU by the hour

Pick a live configuration, choose a template and connect over SSH with your key. Billing stops when you destroy the instance.

Rent A800 PCIE All GPUs