GPU · 125 TFLOPS dense BF16

A10 fine-tuning cost estimates

All sixteen scenarios on the A10: GPU-hours, wall-clock on 8 GPUs and the cost range across 4 providers, at 35% MFU. Prices read 12 November 2026. Estimates, not quotes.

The figure

Where 125 TFLOPS comes from

“BFLOAT16 Tensor Core · 125 teraFLOPS · 250 teraFLOPS* ·”

nvidia.com

Scenarios

Cost by scenario

Scenario	GPU-hours	8 GPUs	Lowest	Median	Highest
7B LoRA fine-tune on 100 million tokens	18	2.2 h	$18	$34	$57
7B LoRA fine-tune on 1 billion tokens	178	22 h	$178	$342	$569
7B full fine-tune on 100 million tokens	27	3.3 h	$27	$51	$85
7B full fine-tune on 1 billion tokens	267	33 h	$267	$513	$853
13B LoRA fine-tune on 100 million tokens	33	4.1 h	$33	$64	$106
13B LoRA fine-tune on 1 billion tokens	330	41 h	$330	$636	$1,057
13B full fine-tune on 100 million tokens	50	6.2 h	$50	$95	$158
13B full fine-tune on 1 billion tokens	495	2.6 days	$495	$953	$1,585
34B LoRA fine-tune on 100 million tokens	86	11 h	$86	$166	$276
34B LoRA fine-tune on 1 billion tokens	863	4.5 days	$863	$1,662	$2,763
34B full fine-tune on 100 million tokens	130	16 h	$130	$249	$414
34B full fine-tune on 1 billion tokens	1,295	6.7 days	$1,295	$2,493	$4,145
70B LoRA fine-tune on 100 million tokens	178	22 h	$178	$342	$569
70B LoRA fine-tune on 1 billion tokens	1,778	9.3 days	$1,778	$3,422	$5,689
70B full fine-tune on 100 million tokens	267	33 h	$267	$513	$853
70B full fine-tune on 1 billion tokens	2,667	13.9 days	$2,667	$5,133	$8,533

Prices used

A10 price by provider

Provider	$/GPU-hr	Configuration	Source
OVHcloud	$1.00	1×	price page
Runcrateours, live	$1.85	1×	price page
Oracle Cloud	$2.00	1×	price page
Microsoft Azure	$3.20	1×	price page

Each provider's lowest listed on-demand configuration for this GPU. Marketplace figures are left out.

Questions

A10: common questions

What is the lowest cost scenario on the A10?

7B LoRA fine-tune on 100 million tokens: about $18 on OVHcloud, 18 GPU-hours.

What throughput is assumed?

125 TFLOPS dense BF16 as printed by the maker, times 35% MFU, so about 44 TFLOPS of useful work.

How many providers list it?

4 with a current provider-set on-demand price. Runcrate is one of them at $1.85 live, marked as ours.

Disclosure

Fine-Tune Cost Lab is published by the Runcrate team, who also run Runcrate, an AI cloud. Runcrate is listed in the table for A10, marked "ours, live". It is sorted by price like every other row, with no pinned position. Other providers' prices are read from their own public pricing pages. We have no paid placement.