Back to browse
ThinkLLM, A knowledge graph of AI models (HTTPS://thinkllm.dev)

ThinkLLM, A knowledge graph of AI models (HTTPS://thinkllm.dev)

by gkanellopoulos·May 23, 2026·1 point·0 comments

AI Analysis

●●SolidSlickSolve My Problem

Hugging Face but organized by use case instead of architecture, with model comparisons.

Strengths
  • Groups 2000+ models by practical tasks like 'Customer Support' rather than technical specs
  • Side-by-side comparison of up to three models with capability matrices
  • LLM-generated descriptions reviewed by humans add context beyond raw metrics
Weaknesses
  • No filtering by license, cost, or latency—critical factors for enterprise decisions
  • Data sourced from public APIs means no unique insights beyond better organization
Category
Target Audience

Enterprise architects, product managers, developers selecting LLMs

Similar To

Hugging Face · Papers With Code · LMSys Org

Post Description

As an Enterprise Architect I work with Capabilities, Use Cases and Value Maps amongst other things. Hugging Face is a great resource for tracking down AI models but is mostly technical and quite detailed. I built ThinkLLM because I thought that as more and more people are going to be using LLMs it would be easier to find AI models by capabilities and use cases than simply browsing long lists of models.

The website has nothing extraordinary or special. Is just a different view on existing data. It basically tries to connect AI models (mainly LLMs) with capabilities, use cases and other useful resources such as glossary terms, research papers and benchmarks. It has also the functionality of comparing up to three models.

The website was built with nextjs and Rust (backend). I used data from publicly available sources. Scoring and model descriptions are generated by using an LLM and reviewed by me prior to publication.

Any comment on how to make it better would help a lot

Thanks

Similar Projects