The use case: how Mirakl deploys Anyscale
Mirakl uses Anyscale's managed Ray platform to run fine-tuned, smaller open-source LLMs with LoRA adapters for batch inference. This automates catalog onboarding for marketplace sellers, with GPU clusters that autoscale to match traffic.
More reported results
Figures published in Anyscale’s case study. They are not independently measured by usedby.
Consistent peak performance, handling 80k+ token/sec generations without latency.
automating onboarding for over 10 million products each month
they can scale up to 20 GPU nodes during peak hours, but scale down to near-zero GPUs overnight
By intelligently routing up to 90% of catalog traffic to LLaMA 3.1 8B
In the customer’s words
Using GPT-4 would have cost us more than half a million dollars per month… with Anyscale and fine-tuned LLaMA models, we brought that down by 3× while scaling to millions of products.Other AI tools used by Mirakl
Other public relationships recorded on usedby.
Explore Mirakl’s AI stackOther companies using Anyscale
Customers across industries, with public evidence.
AttentiveConfirmed
Bedrock RoboticsConfirmed
BonsaiConfirmed
CanvaConfirmed
Chef RoboticsConfirmed
CoactiveConfirmedGo straight to the source
Based on Anyscale’s published case study. We check the public evidence; the results are reported by the vendor.
anyscale.com- Last checked
- Oct 7, 2026
Open source capture What the story claims, and what we checked
What we compared with the page.
- The figure: 3xChecked
Printed word for word on the page, near the name of Mirakl.
- The passage quoted aboveChecked
Copied word for word from the page, near the name of Mirakl.
- Mirakl uses AnyscaleChecked
Confirmed line.
- The result itselfNot checked
We quote it; we did not measure it.
Investing in fine-tuned open-source models and leveraging Anyscale to manage Ray orchestration saved the company 3x in inference costs compared to using OpenAI API for all listings.


