English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Many companies use large language models (LLMs) offered as a service, likeOpenAI's GPT-4, to create AI-enabled product experiences. Along with thebenefits of ease-of-use and shortened time-to-solution, this reliance onproprietary services has downsides in model control, performance reliability,uptime predictability, and cost. At the same time, a flurry of open-sourcesmall language models (SLMs) has been made available for commercial use.However, their readiness to replace existing capabilities remains unclear, anda systematic approach to holistically evaluate these SLMs is not readilyavailable. This paper presents a systematic evaluation methodology and acharacterization of modern open-source SLMs and their trade-offs when replacingproprietary LLMs for a real-world product feature. We have designed SLaM, anopen-source automated analysis tool that enables the quantitative andqualitative testing of product features utilizing arbitrary SLMs. Using SLaM,we examine the quality and performance characteristics of modern SLMs relativeto an existing customer-facing implementation using the OpenAI GPT-4 API.Across 9 SLMs and their 29 variants, we observe that SLMs provide competitiveresults, significant performance consistency improvements, and a cost reductionof 5x~29x when compared to GPT-4.

Scaling Down to Scale Up: A Cost-Benefit Analysis of Replacing OpenAI's LLM with Open Source SLMs in Production