This repository collects machine learning resources and tools for the Ruby programming language.
Open source
24 alternativesEqbench3
EQ-Bench 3 is a benchmark project with an HTML repository and an MIT license.
About Eqbench3
EQ-Bench 3 is an open-source benchmark for testing emotional intelligence in language models. It uses multi-turn role-play scenarios, such as parenting, relationship conflict and workplace tension, and transcript analysis tasks that assess empathy, social reasoning and psychological insight.
A separate judge model, by default Claude Opus 4.6, assigns rubric scores or performs pairwise comparisons to produce ELO-style rankings. Running it requires cloning the GitHub repository, installing dependencies, and configuring API keys for the tested and judge models. The leaderboard is viewable at eqbench.com.
Key features
- Role-play scenarios covering parenting, conflict, workplace tension
- Analysis scenarios interpreting transcripts for human dynamics and subtext
- Rubric scoring 0-20 scaled to 0-100 across empathy criteria
- Pairwise comparison with TrueSkill/ELO-like ranking solver
- Configurable judge model via OpenAI-compatible endpoint, default Claude Opus 4.6
- Local run files mergeable into canonical leaderboard data
Best for: Researchers and developers evaluating emotional intelligence in language models.
Good for
- Open-source repository with MIT license
- Provides both rubric scoring and pairwise ELO ranking
- Judge model can be any LLM via OpenAI-compatible endpoint
Watch out for
- Requires configuring TEST and JUDGE API keys and endpoints
- Requires cloning repository and installing Python dependencies
- Pairwise judgments truncate model outputs
Eqbench3 at a glance
- Website
- github.com/EQ-bench/eqbench3
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 5, 2026 · Online when checked
- Listed
- Oct 5, 2026
Eqbench3 alternatives
Tools and sites you could use instead, ranked by how closely they match.
Browse a list of machine learning models for Apple Core ML, collected in this repository.
Open source
24 alternatives
This repository collects references for vibe coding, where AI helps write code.
Open source
24 alternatives
This repository collects tools and resources for vibe coding with AI assistance.
Open source
24 alternatives
This repository collects prompts, system prompts, and instructions for large language models.
Open source
24 alternatives
A curated list of large language model resources maintained in the Awesome-LLM repository.
Open source
24 alternativesPopular in AI Directories & Resources
Well-known listings from the same topic.
getdesign.mdgetdesign.mdfuturepediafuturepedia.ioAwesomeclaudeawesomeclaude.aiLLM Explorerllm-explorer.comAI Depotaidepot.coAI Valleyaivalley.ai
Eqbench3 FAQ
What judge model does EQ-Bench 3 use?
By default Claude Opus 4.6, but any LLM accessible via an OpenAI-compatible endpoint can serve as judge.
How are models scored?
A judge LLM assigns a multi-criteria rubric score from 0-20, scaled to 0-100, or pairwise-compares transcripts to produce an ELO-style ranking.
Where can the leaderboard be viewed?
The full EQ-Bench leaderboard is viewable at https://eqbench.com/.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.