git404hub

what is bouquet fr?

facebookresearch/bouquet — explained in plain English

Analysis updated 2026-07-17 · repo last pushed 2026-06-15

★ 6PythonAudience · researcherComplexity · 4/5MaintainedSetup · hard

tl;dr

A benchmarking tool that tests machine translation models across 1000+ language pairs and scores their output against a public leaderboard.

vibe map

mindmap
  root((repo))
    What it does
      Benchmarks translation models
      1000 plus language pairs
      Public leaderboard
    Tech stack
      Python
      vLLM
      GPU
    Use cases
      Evaluate translation model
      Compare to leaderboard
      Score with metrics
    Audience
      ML researchers
      NLP engineers
      Model developers

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

what do people make with this?

VIBE 1

Benchmark a translation model locally on GPU or via an OpenAI-compatible API endpoint.

VIBE 2

Score existing translation outputs using ChrF++, MetricX-24, and GlotLID metrics.

VIBE 3

Submit a model's results to the public BOUQuET leaderboard via pull request.

what's the stack?

PythonvLLMGPU

how it stacks up fr

facebookresearch/bouquetaclark4life/home-depot-crawlashishdevasia/ha-proton-drive-backup
Stars666
LanguagePythonPythonPython
Last pushed2026-06-152014-08-10—
MaintenanceMaintainedDormant—
Setup difficultyhardmoderatemoderate
Complexity4/52/52/5
Audienceresearcherdeveloperops devops

Figures from each repo's GitHub metadata at analysis time.

how do i run it?

Difficulty · hard time til it works · 1h+

Requires a GPU for local inference or an API endpoint, plus cluster tooling for parallel runs.

No license information was found in the explanation.

prompts (copy fr)

prompt 1
Help me run this BOUQuET benchmark against my own translation model using vLLM.
prompt 2
Explain what ChrF++, MetricX-24, and GlotLID measure and why all three are used together.
prompt 3
Show me how to plug a custom translation model into this tool's hackable interface.

Frequently asked questions

what is bouquet fr?

A benchmarking tool that tests machine translation models across 1000+ language pairs and scores their output against a public leaderboard.

What language is bouquet written in?

Mainly Python. The stack also includes Python, vLLM, GPU.

Is bouquet actively maintained?

Maintained — commit in last 6 months (last push 2026-06-15).

What license does bouquet use?

No license information was found in the explanation.

How hard is bouquet to set up?

Setup difficulty is rated hard, with roughly 1h+ to a first successful run.

Who is bouquet for?

Mainly researcher.

peek the repo → explain another one

This repo across BitVibe Labs

double-check against the repo, no cap.

similar repos, if this ain't it