M
MercyNews
Home
Back
Sweep lanza modelo de 1.5B parámetros de código abierto para autocompletado de próxima edición
Tecnologia

Sweep lanza modelo de 1.5B parámetros de código abierto para autocompletado de próxima edición

Hacker News10h ago
3 min de lectura
📋

Hechos Clave

  • Sweep ha lanzado un modelo de 1.5B parámetros de pesos abiertos específicamente diseñado para el autocompletado de próxima edición, una función que predice la siguiente edición de un desarrollador basándose en cambios recientes.
  • El modelo está diseñado para ejecutarse localmente en la máquina de un desarrollador, ofreciendo una alternativa que preserva la privacidad frente a los asistentes de codificación basados en la nube, mientras mantiene un alto rendimiento.
  • En pruebas contra modelos como Mercury, Zeta e Instinct, el modelo de Sweep demostró una velocidad y precisión superiores en cinco puntos de referencia diferentes, incluyendo tareas para cambios distantes y autocompletado estándar de código.
  • El proceso de entrenamiento involucró un enfoque de dos etapas: ajuste fino supervisado en 100,000 ejemplos de repositorios con licencias permisivas, seguido de 2,000 pasos de aprendizaje por refuerzo para corregir código que no se analiza sintácticamente y la verbosidad.
  • Un descubrimiento clave durante el desarrollo fue que un formato simple de bloques "original" y "actualizado" era más efectivo para el modelo que los diffs unificados complejos, destacando la importancia de la estructura del prompt para modelos de IA más pequeños.

Una nueva era para el autocompletado de código

El panorama de las herramientas para desarrolladores está cambiando con la introducción de un nuevo y compacto modelo de IA diseñado para predecir el siguiente movimiento de un programador. Sweep, una empresa centrada en el desarrollo asistido por IA, ha lanzado un modelo de 1.5B parámetros entrenado específicamente para el autocompletado de próxima edición. Este enfoque difiere significativamente del autocompletado tradicional de código al analizar el contexto de las ediciones recientes para predecir qué escribirá un desarrollador a continuación.

Lo que distingue a este modelo es su combinación de una huella pequeña y alto rendimiento. Está diseñado para ejecutarse localmente en la máquina de un desarrollador, ofreciendo una alternativa que preserva la privacidad frente a las soluciones basadas en la nube. A pesar de su tamaño, el modelo demuestra capacidades que superan a competidores mucho más grandes, haciendo que el autocompletado avanzado sea accesible sin requerir hardware potente.

Rendimiento y puntos de referencia

La afirmación principal del modelo es su eficiencia excepcional. Es suficientemente pequeño para ejecutarse localmente mientras supera a modelos cuatro veces más grandes tanto en velocidad como en precisión. Para validar estas afirmaciones, los desarrolladores realizaron pruebas rigurosas contra varios modelos establecidos, incluyendo Mercury (Inception), Zeta (Zed) e Instinct (Continue).

La evaluación fue integral, abarcando cinco puntos de referencia distintos diseñados para medir diferentes aspectos de la edición de código:

  • Edición siguiente por encima y por debajo del cursor
  • Funcionalidad de salto con tabulación para cambios distantes
  • Tareas estándar de Relleno en el Medio (FIM)
  • Tolerancia al ruido

A través de esta prueba, surgió una idea clave: se descubrió que la precisión de coincidencia exacta se correlaciona mejor con la usabilidad en el mundo real. Esto se atribuye a la naturaleza precisa del código, donde el espacio de soluciones es relativamente pequeño y los errores son costosos. La capacidad del modelo para predecir la edición exacta siguiente, en lugar de una sugerencia probabilística, se traduce directamente en una experiencia de desarrollador más efectiva.

"El formato verboso es simplemente más fácil de entender para los modelos más pequeños."

— Equipo de Desarrollo de Sweep

La arquitectura de la predicción

La efectividad del modelo no es solo un producto de sus datos de entrenamiento, sino también de su arquitectura subyacente. Un descubrimiento sorprendente durante el desarrollo fue la importancia crítica del formato del prompt. El equipo ejecutó un algoritmo genético sobre 30 formatos de diff diferentes para encontrar la forma más efectiva de presentar los cambios de código al modelo.

El formato ganador resultó ser notablemente simple. En lugar de diffs unificados complejos, el modelo responde mejor a bloques simples de original y actualizado. Este formato verboso y estructurado es más fácil de analizar y entender para el modelo más pequeño, lo que conduce a un mejor rendimiento. El hallazgo subraya que para los modelos de IA, la claridad de la entrada puede ser tan importante como el volumen de datos de entrenamiento.

El formato verboso es simplemente más fácil de entender para los modelos más pequeños.

Entrenamiento y metodología

El modelo se entrenó utilizando un proceso de dos etapas para garantizar tanto un conocimiento amplio como una salida de alta calidad. La fase inicial involucró un Ajuste Fino Supervisado (SFT) en aproximadamente 100,000 ejemplos provenientes de repositorios con licencias permisivas. Esta etapa fue computacionalmente eficiente, requiriendo solo cuatro horas en un clúster de ocho GPUs H100.

La segunda fase, y posiblemente más crítica, utilizó Aprendizaje por Refuerzo (RL) durante 2,000 pasos. Este paso fue diseñado específicamente para abordar casos extremos que el SFT por sí solo no podía resolver. El proceso de RL incorporó dos mecanismos clave:

  • Verificación de análisis con Tree-sitter para asegurar que el código generado sea sintácticamente válido
  • Regularización de tamaño para prevenir salidas excesivamente verbosas

Este enfoque de entrenamiento de dos etapas permite al modelo no solo predecir patrones comunes, sino también generar código que sea tanto analizable como conciso, abordando puntos de falla comunes en la codificación asistida por IA.

Código abierto y accesibilidad

En un movimiento para fomentar la innovación comunitaria, los pesos del modelo se han hecho públicos. La decisión de abrir el código de los pesos está impulsada por el deseo de permitir el desarrollo de herramientas de autocompletado rápidas y que preserven la privacidad para cualquier editor. Este enfoque contrasta con los modelos propietarios que a menudo están bloqueados en plataformas específicas o requieren conectividad a Internet.

El modelo es inmediatamente accesible a través de dos canales principales:

  • Descarga directa desde Hugging Face para integración en proyectos personalizados
  • Un plugin de JetBrains listo para usar para pruebas inmediatas en IDEs populares

Los desarrolladores han invitado explícitamente a la comunidad a construir sobre su trabajo, fomentando contribuciones para otros editores como VSCode y Neovim. Este enfoque abierto podría acelerar la adopción de asistentes de codificación locales y potenciados por IA en todo el ecosistema de desarrolladores.

Viendo hacia adelante

El lanzamiento de este modelo de 1.5B parámetros marca un paso significativo hacia la creación de asistentes de codificación de IA sofisticados más accesibles y eficientes. Al demostrar que un modelo más pequeño y ejecutado localmente puede superar a alternativas más grandes y basadas en la nube, Sweep ha abierto la puerta a una nueva clase de herramientas para desarrolladores que priorizan la velocidad, la privacidad y el control del usuario.

Las conclusiones clave son claras: el futuro del autocompletado de código puede no residir en modelos cada vez más grandes, sino en arquitecturas y metodologías de entrenamiento más inteligentes y eficientes. A medida que la comunidad comience a experimentar con estos pesos abiertos, podemos esperar ver una proliferación de herramientas innovadoras que integren la predicción de próxima edición en una amplia gama de entornos de desarrollo, cambiando fundamentalmente la forma en que los desarrolladores interactúan con su código.

Preguntas Frecuentes

¿Cuál es el desarrollo principal?

Sweep ha desarrollado y liberado como código abierto un modelo de IA de 1.5B parámetros para el autocompletado de próxima edición. Este modelo utiliza el contexto de las ediciones recientes de un desarrollador para predecir su siguiente cambio, con el objetivo de mejorar la eficiencia y precisión en la codificación.

¿Por qué es esto significativo?

El modelo es significativo porque combina un tamaño pequeño y ejecutable localmente con un rendimiento que supera a modelos cuatro veces más grandes. Esto hace que la asistencia de codificación avanzada por IA sea más accesible y amigable con la privacidad, ya que no requiere enviar código a un servidor en la nube.

¿Cómo pueden los desarrolladores usar este modelo?

Los desarrolladores pueden descargar los pesos del modelo directamente desde Hugging Face para integrarlos en sus propios proyectos. Alternativamente, pueden usar el modelo inmediatamente a través del plugin de JetBrains disponible, con el potencial de extensiones construidas por la comunidad para otros editores como VSCode y Neovim.

Continue scrolling for more

La IA transforma la investigación y las demostraciones matemáticas
Technology

La IA transforma la investigación y las demostraciones matemáticas

La inteligencia artificial está pasando de ser una promesa a una realidad en las matemáticas. Los modelos de aprendizaje automático generan teoremas originales, forzando una reevaluación de la investigación y la enseñanza.

Just now
4 min
342
Read Article
Football is the biggest thing in America. Chuck Klosterman says that's going to change.
Sports

Football is the biggest thing in America. Chuck Klosterman says that's going to change.

Big time football — like this month's college football championship — is the most dominant force in American culture. It won't always be that way, writer Chuck Klosterman argues in his new book. Set Number: X164821 TK1 Are you ready for some football? Trick question: America is always ready for more football. It's an appetite without end. But writer Chuck Klosterman, who just devoted a book to the sport, says football's dominance will ultimately be its undoing. The notion that football is the biggest force in American culture and entertainment is so ingrained and obvious that it almost seems like something not worth mentioning. Chuck Klosterman feels otherwise: The pop culture writer just devoted an entire book to the sport, and its meaning and importance. But Klosterman also argues that football won't always be the main thing — and that its overwhelming size and scale will be the thing that eventually undermines it. I talked to Klosterman about all of that, as well as topics like the impact of video games on the sport, and why he thinks paying college football players is good for now, and really damaging in the long run. You can hear our entire conversation on my Channels podcast. What follows are edited excerpts from our chat. Peter Kafka: What is the point of a book called "Football" in 2026? Chuck Klosterman: I have been obsessed with sports and football my entire life. And probably 20 years ago, I made an unspoken, abstract decision — at some point, I want to do a book that's just about sports. My initial idea was that it was going to be about basketball, but I realized that's crazy: If you're writing about something that's part of the culture, football is the sport. It's the only one. If someone said to you, "Explain the last half of the 20th century through some idea, some metaphor," football is the thing to pick. It might not be the case for the 21st century, but it is for the end of the 20th century. And it is for the world we live in right now. One of the big changes in sports — and definitely in football — recently is the legalization of sports betting. It's omnipresent. You seem ambivalent-to-positive about it, which is not where I thought you'd end up. CHUCK: Do I think that gambling in this legalized way is bad for society? I would say probably, for all of the predictable reasons — particularly because it's on your phone and you've put your [financial information] in, so the money does not seem real. Even if it was just a situation where you had to feed dollar bills into it, everything would change. You're taking something that's addictive for some people and marrying it to your phone, which is also addictive for some people. It seems like an obvious way to get in trouble. CHUCK: But for football, it is good. Because it adds a different context for conversation about it. When I have a conversation with other dads, we'll talk about sports a lot, and then there'll be this other conversation. About gambling. We're still technically talking about football, but this is a whole different thing. It has a different context, a different meaning. It says more about the person. Some could argue that it's a weird argument for gambling being good. But if we look at football as a form of entertainment, a distraction, something to consume, to occupy yourself — I think gambling does make it more interesting. This is the pitch from the entire sports gambling industry — it improves the game, it adds stakes, it makes it more interesting. You seem to agree. Certainly there are people who feel that way. I think a lot of people I know who do a lot of gambling, they have a sort of mixed feelings about it — I think it does bother them that suddenly the game seems meaningless if they're not gambling on it. But for me, as somebody who doesn't really put money into it, I find it a fascinating thing. Like: Indiana was an eight-and-a-half point favorite [in Monday's college football championship game]. They did not cover. But the meaning of eight-and-a-half points in a spread suggests that the gambling markets believe it will be a blowout. That's different than saying, "I think Indiana's gonna win easily." I find it to be much more intriguing than I would've thought. I was definitely against the idea of sports betting. Ideologically, I thought, "This is bad." And like I say, socially it probably is. But for each individual sport … You argue that football's grip on America will eventually collapse. How will that happen? Football is not just the most popular thing, or just too big to fail. It is too big to stop expanding. It has to constantly get bigger. The financial demands of it, the amount of revenue it has to make, can only go up. The NFL only operates from the position of "How can we stretch out further? Can we even swallow up all the other sports? Can we expand into Europe and all of these things?" And right now, that's a successful thing. The more they make, the more people consume, the more people spend on it —most importantly, the TV networks and streamers who need it. And the depth of caring is so deep. If football went away this fall for whatever reason, people will be like, "What am I gonna bet on? What is my life gonna be like? Who am I? What's my identity if I'm not a fan of this team?" You really saw this during COVID. "We're still playing college football. No one's in their classes, but we gotta play these games. We can't not play them." So it's really brittle, right? It's a system that has to keep going. But I think at some point there is going to be a change. Probably in advertising. It's not that advertising is going to disappear. It's just that what it costs to buy an ad during a football game … it's not going be worth the trade off. So let's say it gets to be a point when Fox or Amazon Prime, or whoever's carrying these games, renegotiates their NFL contract and for the first time, the number doesn't go up. Maybe even the number goes down. [That puts the] NFL's in this weird position — they have to take their best offer. Then the players will say, "There's no way we're going to take less money. There's no way we're going to start playing 22 games just because you're not making enough money from CBS. We're going to strike." Or the owners will be like, "We're going to lock the players out" [because they can't afford to pay them] — like what's happened with baseball. If that happened now, it would be this American calamity, where people would freak out. But as people have less and less of a personal relationship to the game when this happens in the future, they'll be like, "Well, that's an entertaining distraction. I can replace it with something else." Something has to be part of your life for it to be so important that you'll do whatever you can to keep it going. Why will fans have less of a relationship to football? The comparison I use is horse racing. In the 1920s, along with boxing and baseball, it was the biggest sport. And in the 1920s, the average person still had a real relationship to the culture of horses. They had a blue-collar job, and horses were still doing some of the labor. Or their dad had a horse farm. They definitely saw horses all the time. The horse was part of the world in which they lived. That is no longer the case. Now, horse racing is just for people who own horses and people who gamble on it. That's really all it is. My fear is that football's gonna put itself in a position where it's too big. Its tentacles reach too far. And people will say, "Well, I guess we'll choose something else." And when it collapses, something that size collapses hard. It kind of implodes on itself. I'm not one of these people [who complain about capitalism]. But I do think some of the problems of capitalism are easiest seen through sports leagues, which are smaller simulations of society. And what's happening with the way money operates in pro and college football, it seems precarious to me. The financial side's changing in an exponential way. And when society shifts, it's the big things that can't. They're not nimble. The small things can. Read the original article on Business Insider

11m
3 min
0
Read Article
We're siblings who built a 6-figure Turo business in under 2 years. We started with just our mom's car — here's how we scaled up.
Lifestyle

We're siblings who built a 6-figure Turo business in under 2 years. We started with just our mom's car — here's how we scaled up.

Mike and Darlene Person. Courtesy of JDM Whipz Siblings Michael Rumph and Darlene Person built JDM Whipz, a profitable Turo car rental fleet. JDM Whipz grew to 13 vehicles in Atlanta by reinvesting profits and focusing on customer service. Their experience highlights Turo business strategies, challenges, and the importance of planning. This as-told-to essay is based on conversations with Michael Rumph, 60, and Darlene Person, 61, siblings in metro Atlanta who run JDM Whipz, a 13-car Turo fleet. It has been edited for length and clarity. Darlene: For several years, my brother, Mike, and I talked about wanting to go into business together. We considered a few ideas but hadn't settled on one. In February 2022, we rented an Escalade from Turo for a family trip. Almost exactly a year later, we launched our own Turo business, JDM Whipz. By 2024, JDM Whipz was making six figures in profit. We were fascinated by how quickly and easily we'd rented a car from Turo Darlene: Mike dove into research, sending me videos and websites every week. We began meeting on Sundays to discuss launching a Turo business. By January 2023, we had registered the company, and in late February, we went live on Turo, securing our first booking on the same day. I'm a federal government employee, and I still work full-time. Most of my Turo work takes place in the evenings, when I prep cars after my day job. When we started, both of us were balancing full-time jobs. Mike: I'm an IT nerd with 37 years of experience in desktop support, server support, and project management. Now, I'm a full-time Turo project manager. Our first car was a 2016 Toyota Corolla Mike: The Corolla belonged to our mom, who died in June 2022. It's still on the platform; I just had the front end repainted and protective film added. From there, we purchased a 2017 Mazda, a 2017 Chevy Malibu, a 2017 Equinox, and a 2018 Equinox. By the end of our first year, we had 10 cars. We purchased eight with cash and used "creative financing" for the other two, purchasing them with 0% credit cards Today, we have 13 vehicles — we own 12 outright and have one payment on a 2019 Chrysler Pacifica. We've had five cars totaled in the past year, but because I hold a dealer's license and buy vehicles wholesale at auctions, we've been able to replace them and even come out ahead on insurance payouts. The price difference compared to traditional rental companies is huge Mike: Enterprise once quoted me $805 for an SUV from Friday to Sunday. Turo cost us $360 for four days and 600 miles when we rented that Escalade. Darlene: We listed mom's Corolla as a trial, but the first renter kept extending her trip week after week, and it ended up being 2.5 months total. The car was gone, and we had nothing to do, so we said, "Let's add another." From the start, we reinvested every dollar back into the business. That's how we grew to 10 cars in a year. We choose Turo's 60% protection plan: we keep 60%, Turo takes 40%, but our deductible is lower. Hosts can take up to 90% but risk a much higher deductible. Understanding that trade-off is important. Mike: Day-to-day, I'm checking cars in and out, cleaning, handling inspections, and keeping up with competition in the Atlanta market. I left my IT job last April after paying off all my debts. I was breathing and dreaming of Turo at work. Now it's my focus. Darlene: Atlanta is huge, so Mike covers the north side, and I handle the south. We've never had all 13 cars sitting at our homes at once, where we store them. Usually they're rented out, coming back for a day or two before heading out again. I've learned more about cars in these two and a half years than I ever imagined. Turo does national advertising, but we're also exploring our own marketing efforts Darlene: We're obsessed with customer service. Over the years, we've both complained about poor service elsewhere. When we started JDM Whipz, I told Mike, "I want people to feel valued." Our reviews consistently mention exceptional service, which is the most rewarding feedback we receive. Mike: I built JDMWhipz.com because Turo's search doesn't show business names. Our site lets visitors view our entire fleet in one place without having to search. When ready to book, they tap a link that takes them directly to Turo to complete the reservation process. The biggest challenge for new hosts is capital. Ideally, buy or own your first car outright. Read Turo's Terms of Service carefully to understand the responsibilities of both the host and the guest. Another challenge is time Darlene: Most people start Turo while working full-time. You need excellent planning and time management. We learned the hard way: clean a car as soon as it returns. Early on, we'd say, "I'll do it in the morning," and then get a 7 a.m. booking. At 5 a.m., with car washes closed, you're in the driveway scrubbing, photographing, and prepping the car. Now it's a habit: the car comes in, and it gets cleaned. Mike: Revenue now depends on the season and the vehicle. Our Corolla rents for $42 to $46 per day. Some cars earn $800 a month, while others earn $1,200 during peak season in Atlanta, which typically spans from March to September. We're still making six figures in profit a year. Our biggest advice: do your homework Darlene: Check Turo's website for rules, like no cars older than 12 years or with more than 130,000 miles when listed. Watch videos from Turo hosts on YouTube, but vet their advice. Don't be afraid to ask experienced hosts for guidance. Mike and I mentor people all the time; we love helping new hosts succeed. Our long-term goal is to become Atlanta's go-to personal car rental company. We're proud of what we've built — and even prouder to do it together as siblings. Read the original article on Business Insider

12m
3 min
0
Read Article
Politics

World Cup 2026: Could Europe drive a boycott?

Just as it did in Russia 2018 and Qatar 2022, politics is dominating the build up to World Cup 2026. Could discontent with Donald Trump in Europe see football heavyweights united behind such a cause?

13m
3 min
0
Read Article
Alcalde de Burdeos acepta mediación en disputa por transparencia presupuestaria
Politics

Alcalde de Burdeos acepta mediación en disputa por transparencia presupuestaria

El alcalde de Burdeos, Thomas Cazenave, ha aceptado una mediación judicial para resolver una disputa con la oposición sobre la transparencia presupuestaria y la divulgación de gastos.

15m
5 min
0
Read Article
Instituto de Política Bitcoin, Fedi y Universidad Cornell estudian perspectivas financieras estadounidenses
Cryptocurrency

Instituto de Política Bitcoin, Fedi y Universidad Cornell estudian perspectivas financieras estadounidenses

Una colaboración entre el Instituto de Política Bitcoin, Fedi y la Universidad Cornell estudiará las perspectivas estadounidenses sobre privacidad financiera en un momento de mayor escrutinio regulatorio.

32m
5 min
7
Read Article
Congelados los subsidios, comunidades pagan el precio
Environment

Congelados los subsidios, comunidades pagan el precio

En Sauget, Illinois, un subsidio de $500,000 para monitorear la calidad del aire fue terminado abruptamente. Ahora, los residentes quedan con mala salud y sin prueba del origen.

47m
6 min
12
Read Article
Aplicación no estadounidense lidera tienda danesa en medio de tensiones por Groenlandia
Politics

Aplicación no estadounidense lidera tienda danesa en medio de tensiones por Groenlandia

Una aplicación de boicot a productos estadounidenses ha alcanzado el primer puesto en la App Store de Dinamarca, vinculada a declaraciones políticas sobre el estatus de Groenlandia.

57m
5 min
12
Read Article
¿Qué tan permanente es el asalto de Trump a la acción climática?
Politics

¿Qué tan permanente es el asalto de Trump a la acción climática?

Desde retirarse del Acuerdo de París hasta prohibir la energía eólica offshore, el presidente Trump ha lanzado un asalto integral a la política climática. Sin embargo, muchas de sus medidas son reversibles y su registro legislativo es escaso.

1h
5 min
12
Read Article
Trump anuncia marco para Groenlandia tras disputa con la OTAN
Politics

Trump anuncia marco para Groenlandia tras disputa con la OTAN

Trump anuncia un marco para Groenlandia y abandona amenazas de invasión tras una disputa con la OTAN sobre la creciente actividad china y rusa en el Ártico.

1h
5 min
12
Read Article
🎉

You're all caught up!

Check back later for more stories

Volver al inicio