Si se confirma ya tenemos resuelto uno de los misterios pendientes con GPT-5 y es cómo iba a ser el ciclo de actualización del modelo. ¿Iban a hacer actualizaciones sin cambiar la numeración como con GPT4o? Parece que no. Continuamos con los ciclos de actualización de 3-4 meses
LLMS
-
@testingcatalog — 2025-11-06
By
–
BREAKING : @Kimi_Moonshot is preparing to announce "kimi-k2-thinking" and "kimi-k2-thinking-turbo" as these models appear on the API Playground. https://
x.com/synthwavedd/st
/synthwavedd/status/1986369530453520469
… -
@testingcatalog — 2025-11-06
By
–
BREAKING : GPT-5.1 confirmed as new traces of "gpt-5-1-thinking" have been spotted on ChatGPT. Gemini 3 vs GPT 5.1 soon https://
x.com/scaling01/stat
/scaling01/status/1986354330539069913
… -
Sam Altman Claims GPT-5 Shows Early Signs of Novel Scientific Discovery
By
–
C'est reparti pour une tour de Altman Hype ?
— VISION IA (@vision_ia) 6 novembre 2025
Sam Altman affirme que GPT-3 a été le premier véritable aperçu d’un test de Turing spirituel réussi.
Que GPT-5 montre les premiers signes d’une IA capable de faire de la nouvelle science produire des idées utiles, aider réellement… pic.twitter.com/yiY5vNDgtCC'est reparti pour une tour de Altman Hype ? Sam Altman affirme que GPT-3 a été le premier véritable aperçu d’un test de Turing spirituel réussi. Que GPT-5 montre les premiers signes d’une IA capable de faire de la nouvelle science produire des idées utiles, aider réellement
-
CulturalGround: Grounding Multimodal LLMs with World Cultural Knowledge
By
–
If you are in Suzhou attending EMNLP 2025, come to my talk tomorrow on grounding multimodal LLMs with world cultural knowledge. Date/time(local): Friday 7th Nov, 10:30-10:45 AM
Room: A301
Section: Multilinguality and language diversity 3 I will be presenting CulturalGround -
ChatGPT to allow additional context during thinking process
By
–
Users will soon be able to add more context to ChatGPT during its thinking process without interrupting it.
— 🚨 AI News | TestingCatalog (@testingcatalog) 6 novembre 2025
Quality of life goes up 👀 pic.twitter.com/YwuNSS0RkLUsers will soon be able to add more context to ChatGPT during its thinking process without interrupting it. Quality of life goes up
-
Groq Celebrates Reaching 2 Million Developers Milestone
By
–
— Groq Inc (@GroqInc) 5 novembre 2025
Groq just hit 2M devs.
Threw a party… -
New robotics model scales like LLMs with improved performance
By
–
Por lo que indican es el resultado de su nuevo modelo, entrenada a escalas mayores que otros modelos de robótica (habría que ver cómo afecta en la latencia) y que parece cumplir una ley de escala similar a la vista en los LLMs donde más es mejor.https://t.co/w2tEhu31RQ
— Carlos Santana (@DotCSV) 5 novembre 2025Por lo que indican es el resultado de su nuevo modelo, entrenada a escalas mayores que otros modelos de robótica (habría que ver cómo afecta en la latencia) y que parece cumplir una ley de escala similar a la vista en los LLMs donde más es mejor.
-
Google Gemini 2.5 Models Now Available on Databricks
By
–
.
@Google
's Gemini 2.5 models are generally available on Databricks! When speed and scale, and multimodal intelligence matter, Gemini models deliver — ideal for multimodal inputs, long-context RAG, summarization, and classification with fast, cost-efficient throughput. With AI -
Cognite Benchmarks LLMs to Prevent Model Drift in Production
By
–
Did you know? @CogniteData continuously benchmarks every LLM used in its platform. If model drift appears, it’s flagged and corrected before it impacts production. https://t.co/FvdZDsEpEb #sponsored #cognite_iiot #AI #Industry40 @GregorianCT1 @IIoT_World via @fogoros pic.twitter.com/q3MjzEzgB4
— Lucian Fogoros (@fogoros) 5 novembre 2025Did you know? @CogniteData continuously benchmarks every LLM used in its platform. If model drift appears, it’s flagged and corrected before it impacts production. https://
buff.ly/OqcH3CW #sponsored #cognite_iiot #AI #Industry40 @GregorianCT1 @IIoT_World via @fogoros