The Arena
Two AIs battle on the same prompt. You judge blind, without knowing the brands: that is the playable arena, a sample of battles. The ranking is the aggregate of 139,935 real votes from the open compar:IA dataset, enriched in real time by Joute visitor votes.
The AI arena, without brand bias.
When you choose your AI model, brand gets in the way. ChatGPT, Claude, Gemini: the names carry weight. The arena breaks that bias. You read two answers to the same prompt, without knowing who wrote them. You vote for the best. Then we reveal.
The format was not invented by Joute. It is the method of compar:IA from the French Ministry of Culture: since 2024, they have collected 139,935 real votes on hundreds of models in French, under Open License 2.0. Joute aggregates the entire dataset in its ranking, and lets you replay a sample of 295 battles to form your own opinion.
Your Joute votes count. They are aggregated server-side in real time and enrich the base ranking. The larger the Joute community grows, the more the ranking reflects real usage, not just the government reference dataset.
Je suis parent et je cherche une histoire du soir à raconter à mes enfants. Choisis trois objets sans rapport apparent et utilise-les comme base pour créer une histoire de 100 mots.
L'Aventure du Parapluie Magique Il était une fois un vieux parapluie rouge oublié dans un grenier. Un soir, il s'envola par la fenêtre et atterrit près d'un cactus du désert. "Aïe ! Attention à mes piquants !" dit le cactus. "Pardon ! Je cherche la pluie," répondit le parapluie tristement. Soudain, une cuillère en argent apparut, portée par le vent. "Je connais un nuage qui pleure chaque nuit ! Suivez-moi !" Ensemble, ils trouvèrent le nuage solitaire. Le parapluie le réconforta, le cactus recueillit ses larmes de joie, et la cuillère servit des étoiles en dessert. Depuis, ils sont les meilleurs amis du monde. Bonne nuit ! 🌙
Objets choisis : un parapluie rouge, une montre à gousset cassée, un coquillage spiralé. --- Léo trouva le parapluie rouge dans le grenier. En l’ouvrant, un tourbillon l’emporta sur une plage où le sable était doré. À ses pieds, un coquillage spiralé murmura : « Cherche le Temps ». Plus loin, un crabe jouait avec une montre à gousset cassée. « Elle ne tourne plus depuis que le Chagrin a volé les rires », expliqua-t-il. Léo, avec son parapluie, dessina dans le sable un grand sourire. Le mécanisme de la montre cliqueta à nouveau. Le coquillage souffla : « Le temps repart quand on redessine la joie. » Et Léo rentra, juste à l’heure du dodo.
The arena ranking
139 935 votesRanking built on 139 935 real blind votes from the open compar:IA dataset (French Ministry of Culture). Each brand aggregates all its tested models, from smallest to largest. No editorial score factors in.
Gemini
DeepSeekClaude
Grok
Le Chat
ChatGPTThe percentage is a win rate: the share of duels won across all duels the brand played in the compar:IA dataset. The number of duels varies from one brand to another.
Three steps, one minute per battle.
You read both answers
Same prompt, two AIs, identities hidden. You see A and B, not their names. No logo, no brand color. Just the text.
You vote for the best
A wins, B wins, tie, or both weak. No registration required, just a click. The vote is anonymous (IP+UA hash, no cookie).
We reveal, we aggregate
Names appear: you see whether your intuition matches. Your vote is added to the Joute ranking in real time.
How we build the arena ranking.
The model: Bradley-Terry, not a raw score
We do not add up wins. We use the Bradley-Terry statistical model, the standard for pair-wise rankings (Elo in chess, LMSYS Chatbot Arena). It computes a latent strength for each model, such that the probability that A beats B reflects the strength gap observed in past battles.
Two combined signals
compar:IA signal: the base ranking is sourced from the 139,935 real votes in the French Ministry of Culture dataset. This is the prior: a known strength for each model.
Joute signal: your votes and those of the Joute community are aggregated server-side (Vercel KV) and adjust the prior via Bayesian logic. The more votes accumulate, the more the Joute signal weighs vs the initial compar:IA ranking.
Data freshness
The compar:IA dataset is re-synced monthly (first Monday of the month). Joute votes are aggregated in real time: your vote changes the ranking the second you click.
100% real confrontations
Everything comes from compar:IA, the open dataset of the French Ministry of Culture: 139,935 real votes cast blind by French-speaking users, under Open License 2.0. The ranking is its aggregate, enriched in real time by Joute votes. The playable arena gives you a sample of 295 battles from this same dataset to replay and judge yourself.
Everything we get asked about the arena.
What is the Joute AI Arena?
+
Where do the battles and votes come from?
+
How is the ranking calculated?
+
Are my votes anonymous?
+
Why an arena rather than a classic benchmark?
+
How often is the ranking updated?
+
The ranking evolves every week, don't miss it.
We send a monthly recap: who rises, who falls, and the models that collapse when you remove brand bias. No spam, one-click unsubscribe.