back_office_ops · ecommerce · workflow
eBay adapts Meta Llama 3.1 to the e-commerce domain via continued pretraining, achieving approximately 25% benchmark improvement
General-purpose LLMs such as GPT-4 and Claude were too costly for eBay's at-scale needs and introduced data security risks by relying on third-party providers, while training a domain-specific LLM entirely from scratch was too time- and resource-intensive.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · E-commerce data gathering and filtering
eBay gathers data from public listings and product reviews on the eBay website, then thoroughly filters and serializes it.
Tools used
Llama 3.1Megatron-LMNVIDIA H100 80GB GPUNVIDIA NVLinkInfiniBandflash-attention-2
Outcome
The final e-Llama models demonstrate approximately 25% improvement in English e-commerce benchmarks and about 30% for non-English, with only 1% degradation on general domain NLU benchmarks, enabling eBay to drive new AI initiatives across the company.
What failed first
Third-party LLMs (GPT-4 and Claude) were prohibitively expensive and introduced data security constraints, while off-the-shelf open models such as Llama 3.1 lacked the e-commerce domain knowledge required for eBay's use cases.
Results
Time savedaround one month
Volumeapproximately 25%
Grounding & classification
Source type: technical build writeup
21 fields verified against source quotes.
content generationproduct catalogmetric backednamed customertools describedworkflow describedecommerceaccuracy improvementcost reductiontechnical build writeupback office ops