jev-bench · House #24,895

What it is

Calls itself “benchmark”.

• Calls itself: benchmark

“…jev-bench Open-source Jev benchmark: tested against frontier LLMs”

“Open-source Jev benchmark: tested against frontier LLMs”

Who it sells to

Sells to developers.

• Sells to: developers

“An independent, reproducible benchmark of TypeSafe's Jev against openai/gpt-6-luna (cheap LLM) and openai/gpt-6-astra (frontier LLM).”

How it runs

A command-line tool. Runs locally and works offline. Distributed through the Github. Offers an API. Open source. AI is the core of the product. You can bring your own API key.

• Kind: cli
• Runs where: on the device, offline
• Distributed via: Github
• API: yes
• Open source: yes
• AI role: core to the product
• Bring your own key: yes

“All code is open source — run it yourself and verify the results.”

“Github -> https://www.producthunt.com/r/JWHTK5F34JJSEC?utm_campaign=producthunt-api&utm_medium=api-v2&utm_source=Application%3A+Tablif+%28ID%3A+291874%29”

What it works with

Names GitHub Copilot, OpenRouter and 3 more on its pages.

• Names: GitHub Copilot · OpenRouter · TypeSafe's Jev · +2

“GitHub Copilot Write better code with AI”

“All models were called through OpenRouter : Jev via the Decisions API”

What it offers as proof

Live, no waitlist. Publishes docs.

“You must be signed in to change notification settings”

“README.md -> https://github.com/PavelRavvich/jev-bench/blob/main/README.md”