← Latest products

New Benchmark from SWE-bench team is 0% solved

ProgramBench evaluates whether language models can rebuild programs from scratch.

Visit programbench.comexternal site

Reactions (1)

Similar products

by description similarity

  1. Cheddar-bench

    github.com

    unsupervised benchmark for coding agents

    Hacker News9 pts

  2. Mandelbrot

    easylang.dev

    programmed in widely unknown Easylang, speeded up by WASM

    Hacker News22 pts

  3. SOCBench

    github.com

    an open benchmark for AI on SoC tasks

    Hacker News6 pts

  4. New eval from SWE-bench team evalutes LMs based on goals not tickets

    codeclash.ai

    CodeClash: Benchmarking Goal-Oriented Software Engineering

    Hacker News5 pts

  5. Benchmark

    swesweep.com

    AI doesn't find bugs unless you tell it what's wrong

    Hacker News5 pts