Kimi K3.1 Pro de Moonshot AI ofrece un contexto masivo de 1M de tokens y un rendimiento competitivo a menores costos, desafiando a los principales modelos propietarios mientras promete pesos abiertos a finales de 2026.
Resumen generado con IA. Consulta el original para comprobar el contexto y la precisión.
Este es el extracto disponible del feed. Puede estar abreviado o en el idioma original.
Moonshot AI has just released Kimi K3.1 Pro, the first model in its K3.1 series. Kimi K3.1 Pro is a 3T-class open-weight* model with a 1-million-token context window. Moonshot team positions it against the strongest proprietary models on coding, agentic and knowledge benchmarks. For developers, it is cheaper API pricing than K3 and open weights yet not available. *open weights are expected to release in November 2026. TL;DR Size: 2.8T total, 104B active parameters (MoE), 1M-token context. Runs on: Kimi API today. For K3, Moonshot recommends supernodes of 64+ accelerators, with MXFP4 weights and MXFP8 activations. Not a single-GPU model. Performance: Near the proprietary frontier on terminal coding; behind it on scientific terminal tasks. Best: SWE-Marathon 51.7, ahead of every model in Kimi’s chart (self-reported). Worst: Terminal-Bench Science 0.1 at 38.1, versus 68.1 for GPT-6 Astra in Kimi’s chart. What is Kimi K3.1 Pro? Kimi K3.1 Pro is a post-trained version of Kimi K3 with much larger post-training compute. Moonshot team ran joint reinforcement learning across hundreds of downstream domains. To support this, the Moonshot team optimized the RL framework, long-context inference, sandbox infrastructure and rollout scheduling. The base K3 architecture uses Kimi Delta Attention (KDA) and Attention Residuals (AttnRes). It activates 16 of 896 experts through a Stable LatentMoE framework. The model accepts text, images and video through the API, per the quickstart docs. Moonshot team also shipped Kimi K3.1 Pro Highspeed. According to the official Moonshot blog, it runs 2 to 3x faster. How does Kimi K3.1 Pro perform on benchmarks? All numbers below are self-reported by Moonshot at max reasoning effort, mostly with the Claude Code harness. Coding: On Terminal-Bench 4.0, K3.1 Pro scores 65.7%. That is just behind Claude Opus 5.5 at 66.4% in Kimi’s chart. On SWE-Marathon it leads at 51.7, ahead of Opus 5.5 at 46.3. On Kimi Code Bench 3.0 Long-horizon, an int
Una ayuda automática para ordenar, basada en fuentes, palabras clave y entidades, en una escala de 1–12. Alta: 9–12; media: 6–8; baja: 1–5. No mide urgencia, verifica noticias ni representa respaldo editorial.
Los autoencoders CADRE de Kyutai reducen la latencia de los modelos de mundo a menos de 230 ms, lo que permite simulaciones interactivas más rápidas y responsivas en una sola GPU H100.
La conferencia de Martha Minow en el MIT destaca cómo la IA amenaza los cimientos de la democracia, al tiempo que ofrece herramientas potenciales para reforzar la gobernanza de EE. UU., instando a la acción para sostener el constitucionalismo.
El simposio de SHASS del MIT destacó la importancia perdurable de la comprensión humana y las contribuciones de la escuela, en un momento en que la inteligencia artificial transforma cada vez más la educación, la investigación y la sociedad.