Google annonce Gemini 3.6 Flash et une IA de cybersecurite, tout en evoquant 3.5 Pro et Gemini 4
Validez la version 3.6 Flash sur vos bancs de test de code avant de mettre a jour vos pipelines de production.
Pourquoi c'est important
Cette iteration rapide montre l'arbitrage permanent des fournisseurs entre optimisation du cout des tokens et qualite de generation. La gestion de l'obsolescence prematuree des modeles devient un enjeu d'architecture logiciel.
Resume
Apres les retours qualifies de mitigés sur les capacités de codage de Gemini 3.5 Flash, Google ajuste sa gamme d'operateurs legers. La firme renforce simultanement sa verticale securite tout en modifiant le calendrier de ses modeles haut de gamme.
Impact concret
Les equipes d'ingenierie doivent basculer leurs integrants existants vers la version 3.6 Flash suite a la deprecation directe de la mouture precedente.
Applications concrètes
Generation de code economique pour scripts du quotidien et analyse automatisee de menaces informatiques.
Texte source original
Google announced a significant evolution of its AI models at I/O in May with the release of Gemini 3.5 Flash, and it's not slowing down. The company revealed three new AI models today , including its first version of Gemini geared toward cybersecurity. However, none of the new models is the delayed Gemini 3.5 Pro, which was supposed to launch in June. Gemini 3.5 Flash, which was the star of the show at I/O , has already been deprecated. In its place, developers and users will find Gemini 3.6 Flash. Google makes the usual claims about this model—it's marginally more capable and better at coding, and it has great multimodal features. Google says the changes to 3.6 Flash were made in response to user feedback on the 3.5 release. In general, Gemini 3.5 Flash didn't appear to live up to Google's promises around code generation. Perhaps that is simply a consequence of Google's intense focus on efficiency as businesses have started to fret over the cost of AI tokens. Read full article Comments