InnerVault

nagi-bench

IndexedGitHub, nagi-studio

One-shot LLM eval cases by NAGI STUDIO - same prompt, different agents (model + harness), runnable artifacts side by side.

Open the source (opens github.com) https://github.com/nagi-studio/nagi-bench

What it is

Datasets & benchmarks from nagi-studio.

The problem it solves and who it suits haven't been reviewed by InnerVault. The description above is the source's own.

How to get it

Get the code

git clone https://github.com/nagi-studio/nagi-bench.git

Standard git access to a public repository

Setup steps: not provided by the source. Follow the instructions on the source.

Related

  1. bigcodebench

    Datasets & benchmarks

    [ICLR'25] BigCodeBench: Benchmarking Code Generation Towards AGI

    Stale Archived GitHub, bigcode-project 516 ★Apache-2.0Updated 10 months ago 1 note

    git clone https://github.com/bigcode-project/bigcodebench.git
  2. geobench

    Datasets & benchmarks

    GeoGuessr benchmark for language models

    Indexed GitHub, ccmdi 65 ★MITUpdated 3 days ago

    git clone https://github.com/ccmdi/geobench.git
  3. confabulations

    Datasets & benchmarks

    Hallucinations (Confabulations) Document-Based Benchmark for RAG. Includes human-verified questions and answers.

    Stale GitHub, lechmazur 249 ★Updated 15 months ago

    git clone https://github.com/lechmazur/confabulations.git
  4. nyt-connections

    Datasets & benchmarks

    Benchmark that evaluates LLMs using 759 NYT Connections puzzles extended with extra trick words

    Indexed GitHub, lechmazur 250 ★Updated 5 days ago

    git clone https://github.com/lechmazur/nyt-connections.git
  5. prism

    Datasets & benchmarks

    Открытый бенчмарк LLM: какая нейросеть лучше пишет код 1С:Предприятие (BSL). Объективная оценка LLM по методике SMOP с реальным исполнением в 1С — Claude, GPT, Gemini, DeepSeek, YandexGPT, GigaChat.

    Stale GitHub, genlab-1c 55 ★MITUpdated 18 days ago

    git clone https://github.com/genlab-1c/prism.git
  6. debate

    Datasets & benchmarks

    Adversarial multi-turn benchmark for LLM debate quality, using side-swapped matchups and multi-model judging to rank models by judged debate performance.

    Stale GitHub, lechmazur 38 ★Updated 21 days ago

    git clone https://github.com/lechmazur/debate.git