Noticias
ranking top Internet services in 2024
Since the late 1990s, millions have relied on the Internet for searching, communicating, shopping, and working, though 2.6 billion people (about 31% of the global population) still lack Internet access. Over the years, use of the Internet has evolved from email and static sites to social media, streaming, e-commerce, cloud tools, and more recently AI chatbots, reflecting its constant adaptation to users’ needs. This post explores how people interacted online in 2024, based on Cloudflare’s observations and a review of the year’s DNS trends.
Building on similar reports we’ve done over the past several years, we have compiled a ranking of the top Internet properties of 2024, with the same categories included in 2023, including Generative AI. In addition to our overall ranking, we chose 9 categories to focus on:
-
Generative AI
-
Social Media
-
Ecommerce
-
Video Streaming
-
News
-
Messaging
-
Metaverse & Gaming
-
Financial Services
-
Cryptocurrency Services
As we have done since 2022, our analysis uses anonymized DNS query data from our 1.1.1.1 public DNS resolver, used by millions globally. We aggregate domains for each service (e.g., twitter.com, t.co, and x.com for X) and identify the sites that provide services to humans, thus excluding technical domains like root-servers.net. Rankings reflect relative popularity within categories, not absolute traffic. Therefore, a drop in rank doesn’t always indicate less traffic to a specific Internet service — it may simply reflect increased competition from other services, leading to a change in rank.
This part of the 2024 Cloudflare Radar Year in Review highlights shifts in Internet services, with rising platforms like Temu, GitHub Copilot, and WeChat reflecting changing user preferences. ChatGPT (OpenAI) also played a more prominent role in the generative AI space and in our Overall ranking, nearly reaching the Top 50. Major events like the Paris Olympics and US elections influenced rankings as well, boosting Olympics-related sites and news platforms like CNN and Fox News.
Keep reading for a detailed look at the evolution of trends throughout the year. For more, visit our 2024 Cloudflare Radar Year in Review microsite. Along with the lists of most popular Internet services, the Year in Review microsite and its associated blog post explore a number of additional metrics.
Google is again #1. Facebook, Apple and TikTok follow
Since 2021, we’ve started our review of rankings with an Overall Top 10 list, showcasing the most popular Internet services globally based on DNS traffic from our 1.1.1.1 resolver. Unsurprisingly, Google (including services like Google Maps and Google Calendar) remained the #1 Internet service in 2024. Since introducing our ranking method two years ago, no other service has come close to challenging Google’s top spot. It’s important to note that Apple and Microsoft are similar to Google in that their main domains (apple.com or microsoft.com) are used for many different services. We include other services separately, such as Outlook or iCloud, which use their own specific domains.
Top 10 most popular Internet services in 2024, overall
-
Google
-
Facebook
-
Apple
-
TikTok
-
Amazon Web Services
-
Microsoft
-
Instagram
-
YouTube
-
Amazon
-
WhatsApp
Beyond Google, Facebook consistently held the #2 spot throughout 2024. Last year, it competed with Apple for that position. Apple, which uses domains like apple.com for services related to its software and devices, was generally #3. However, TikTok challenged that position on several days since late August. Amazon Web Services (AWS), differentiated from Amazon by domains like amazonaws.com, performed better this year compared to 2023. It held the #5 spot but often traded places with Microsoft during the year.
Instagram also rose in the rankings. It was around #8 in 2023 and steadily improved. Now, it holds the #7 spot, ahead of YouTube.
Amazon remained at #9 for most of the year, the same as in 2023. WhatsApp, owned by Meta, appeared in the Top 10 for the first time, taking the #10 spot.
Close to the Top 10 were Apple’s iCloud, Netflix (which performs better on weekends), and Microsoft’s Outlook.
In the chart below, you can follow the evolution of the top Internet services in our Overall ranking throughout the year.
In 2022, X (then known as Twitter) ranked as high as #10 in our overall ranking and was close to Instagram. It never reached the top 10 in 2023, and in 2024, X dropped further, to #14 or #15. More on X’s performance in the Social Media category below.
Ready to face the Generative AI era?
Generative AI gained global attention in late 2022 with the launch of ChatGPT, and became a global phenomenon during 2023. By 2024, ChatGPT (OpenAI) continues to be by far the most popular service in this category, which includes chatbots, coding bots, and more. Other generative AI services had more stable rankings compared to 2023.
Top 10 Generative AI services in 2024
-
ChatGPT (OpenAI)
-
Character.AI
-
Codeium
-
QuillBot
-
Claude (Anthropic)
-
Perplexity
-
GitHub Copilot
-
Wordtune
-
Poe
-
Tabnine
Significant changes occurred below ChatGPT’s first place ranking throughout the year. Character.AI, an AI-driven chatbot platform, maintained a strong #2 position, staying ahead of Codeium, a code-generation AI tool that has improved its position since June, and Quillbot, an AI writing and paraphrasing tool.
Claude, the AI chatbot from Anthropic, rose in the rankings, particularly after March 4, when the new model, Claude 3, was introduced, and again later in May when it became available in Europe. It reached #5 in June. Perplexity, an AI-driven search and Q&A platform, started the year outside the Top 10 but ended close to Claude. It surpassed Claude for the first time on November 6, 2024, the day after the U.S. elections, reaching #6.
This next chart shows movement among the Generative AI services that were more popular later in the year.
GitHub Copilot’s rise to the Top 10
Several new players entered the Top 10 AI rankings in 2024, showing strong growth. GitHub Copilot, an AI-powered coding assistant, experienced the fastest rise, entering the Top 10 in September (after reaching the Top 20 in June) and staying mostly between #5 and #3 by November, as the next chart shows. Similarly, Suno AI, an AI-powered audio and music generation platform, entered the Top 10 in April, briefly dropped out, but stabilized between #6 and #10 after October — in November, it ranked #6 on weekends.
Some platforms lost ground in the rankings. Wordtune, an AI writing assistant, peaked at #4 during mid-year but declined afterward. Tabnine, another AI-powered coding assistant, held the #5 spot for months but slipped after July. In contrast, Sider AI, a coding assistant, entered the Top 20 in March and finished the year around #12. Poe, an AI chatbot platform, ranked #5 in 2023 and between #5 and #6 before June, but ended 2024 moving around #10, performing better during weekends.
Google Gemini, Google’s AI assistant and model, performed better on weekdays and started the year ranking between #7 and #10, but dropped out of the Top 10 after July as newer AI platforms gained momentum. Hugging Face, an open-source AI and machine learning platform, mostly fluctuated between #7 and #9 during the year, peaking at #4 on August 18 around the time several models were updated, and and as it reached its milestone of 5 million users. However, it fell out of the Top 10 by September.
Midjourney, an AI-powered platform for generating images, performed well until June, when it was close to the Top 10. Additionally, the OpenAI API ranked #18 in the Generative AI category on May 14-15, coinciding with OpenAI’s announcement of GPT-4o availability, including in the API.
ChatGPT’s growth to the Top 50 of our Overall category
Notable trends that we observed when looking at trends for Generative AI services within our larger Overall ranking include:
-
ChatGPT continued its growth in 2024, similar to 2023. In early 2023, it ranked around #200 and ended the year near the top 100. In 2024, it started close to the top 100, reached the top 60 in May with the release of the 4o model, and has been near the top 50 since September, aligning with the return of workers and students to their routines. It ranks higher on weekdays, averaging #56, and drops on weekends.
-
Comparing ChatGPT with other known and non-AI related websites, by late November, ChatGPT ranked ahead of Weather.com, Temu, eBay, Telegram, Google Calendar, and Prime Video, but trailed Disney Plus
Character.ai also showed a clear growth trend in our Overall ranking, from outside the top 200 earlier in the year, to above #180 after July, performing better in August, reaching as high as #161. The AI-driven chatbot platform performed better on weekends than on weekdays, the opposite of ChatGPT.
According to Kepios, there are an estimated 5.22 billion social media users worldwide in 2024 (up from 4.95 billion last year), representing 63.8% of the global population. Social media continues to play a major role in daily life, serving as a key platform for communication, information, and attention.
Once again, social media giants like Facebook, TikTok, and Instagram dominate, ranking among the top 10 most popular Internet services overall.
Top 10 Social Media services in 2024
-
Facebook
-
TikTok
-
Instagram
-
X
-
Snapchat
-
LinkedIn
-
Discord
-
Kwai
-
Pinterest
-
Reddit
In the Social Media category rankings, the top seven remain unchanged from last year. However, there are notable developments in this category. In 2022, X briefly challenged Instagram for the #3 spot during a few days. Since 2023, X has held a solid #4 position, with Snapchat closing in and reaching #4 for the first time on several days in September and October.
LinkedIn stayed steady at #6, followed by Discord. Kwai, a Chinese video app popular in Brazil (with 60 million reported users) and other countries (a reported ), rose from #10 last year to #8. Further down the list, Pinterest kept its #9 rank, while Reddit, previously #8 in 2023, dropped to #10 this year, but peaked at #7 on November 26, just before Black Friday and Thanksgiving in the US. Here’s the Social Media Top 10 chart for 2024:
Our global ranking also highlights several non-Western platforms in the Top 20. These include Douyin (#11), the Chinese version of TikTok; VK (#12), often referred to as the Russian Facebook; and TikTok rivals popular in Southeast Asia SnackVideo (#13) by Chinese Kuaishou (that also owns Kwai). OnlyFans appeared consistently in the Top 20 starting in September, ranking around #18 and surpassing Tumblr by late November.
The #18 spot was briefly held by X alternative Threads (by Instagram) in late September and by Bluesky starting November 18. Mastodon-related servers reached as high as #19 for several days since late August. Here’s a look at X (on top) and its alternatives in this category:
Alternatives to X: Bluesky, Threads, and Mastodon
Let’s move beyond the Social Media category to see how these platforms performed in our Overall ranking, where bigger shifts between services are evident.
As we’ve seen, Threads, Bluesky, and Mastodon (via an aggregation of popular servers) didn’t break into the Top 10 of the Social Media category. However, in the Overall ranking, Mastodon servers, bundled together, consistently ranked between #208 and #248, performing better on weekends.
Bluesky entered the Top 250 in September 2024, and gained additional attention after the US elections. It rose sharply after November 14, peaking at #193 on November 20, and has since stabilized around #220.
Threads entered the Top 250 in August 2024, peaking at #183 on September 24 before dropping out in October. In 2023, Threads peaked at #227 in early July but fell out of the Top 250 by late August. It’s worth noting that Threads also uses Instagram’s cdninstagram.com for images and videos, which may influence Threads position in our DNS rankings (that said, Instagram wasn’t impacted by Threads appearance in our rankings).
Here are some other trends we observed among social media apps, and how they did in our Overall ranking:
-
Instagram’s best day (#6 in the Overall ranking) was August 5, 2024, coinciding with the week the app was banned in Turkey.
-
X’s best day of the year in our ranking was April 14, when it reached #12. This coincided with Arsenal losing the top position in the English football/soccer Premier League (the most-watched sports league in the world) to Manchester City, which went on to win its fourth title in a row. Last year, we noted how football/soccer in England impacted X’s ranking. X also reached #13 on August 9 and 10, during the final weekend of the Paris 2024 Olympics.
-
X performed better on weekdays, while LinkedIn ranked higher between Mondays and Wednesdays. Snapchat and Discord performed better on weekends.
-
Reddit consistently stayed in our Top 50 in 2024, showing growth from around #45 to #40 by November, with a peak at #38 on November 26. It performed better between Mondays and Wednesdays.
-
Quora displayed a downward trend in our ranking, dropping from around #140 to #160. It performed better between Mondays and Wednesdays.
-
Tinder, which performs better on Sundays, started the year around #150 but eventually dropped below #160.
-
Tumblr followed a similar pattern, dropping out of the Top 200, where it was in early 2024, to outside the ranking entirely since September. Tumblr performed better on weekends.
-
OnlyFans showed growth in our Overall ranking, sitting around the Top 220 with a peak at #213 on December 1. It performed better on weekends.
E-commerce: Temu means growth
The importance of e-commerce continues to grow, as highlighted in our recent Cyber Week 2024 blog post. Amazon leads the category, followed by Taobao, the Chinese marketplace, holding a steady #2 spot as it also did in 2023. New to #3 is AliExpress, the global online retail giant from China.
Top 10 E-commerce services in 2024
-
Amazon
-
Taobao
-
AliExpress
-
Shopify
-
Temu
-
Alibaba
-
eBay
-
Shein
-
Mercado Libre
-
Wildberries (RU)
Compared to 2023, eBay lost its #3 spot globally and dropped down to #7, despite starting 2024 at #3 for several days. AliExpress claimed #3, followed by Shopify (#4), the Canadian platform hosting numerous online stores, and Temu (#5). Temu, the low-cost, fast-fashion marketplace launched in the US in September 2022, ended 2023 at #7 but rose to #5 in 2024, occasionally reaching #4 since August. Alibaba dropped to #6 in September.
Shein, the Chinese fast-fashion brand, continued its growth and overtook Mercado Libre (#8) in November. A surprise this year was Wildberries, often called Russia’s “Amazon,” that has been expanding to several neighboring countries (including some in Europe). It climbed to #10 in September, surpassing OLX (which held #10 for several months), Rakuten, and Lazada.
The Black Friday overall effect
Looking at how e-commerce sites performed in our Overall ranking, we observed the following trends:
-
Amazon fluctuated between #9 and #10 after October, returning to #9 on November 30 and December 1, during the Black Friday weekend. It often performed better on weekends.
-
Shopify’s best day of the year was Black Friday, November 29, when it reached #55. The global e-commerce platform performed better during weekdays.
-
Temu, known for low-cost products, started 2024 outside the Top 100 but climbed into the Top 70 by year-end. It performed best in late October and early November, peaking at #63, with a Black Friday spike to #65.
-
Shein, the Chinese fast-fashion brand, showed growth, nearing the Top 100 in early 2024 before dropping to the Top 140 between June and October. It rebounded in November, peaking at #83 on Black Friday. A similar trend was observed in 2023, when it ended the year around the Top 120. Here’s the comparison between recent players Temu and Shein:
-
eBay consistently ranked between #72 and #80, peaking at #62 on October 5-6 and again in late November, just before Black Friday. It often performed better on weekends.
-
Mercado Libre, the Latin American marketplace, had its best day on Black Friday, November 29, reaching #100.
-
Adidas entered the Top 250, ranking #232 on Black Friday, November 29.
-
Target performed well in November, peaking at #133 on November 27, the day before Thanksgiving in the US, and at #127 on December 1. It often performed better on Sundays.
-
Walmart improved its performance from September onward, with its best days on November 25-26, reaching #150.
-
Ikea, the Swedish furniture retailer, peaked at #247 on June 29.
Video streaming: YouTube and Netflix remain uncontested leaders
The relevance of video streaming platforms shows no signs of fading. In 2024, the Top 3 rankings stayed unchanged from 2023, with YouTube firmly holding the #1 spot, followed by Netflix. Among paid streaming services, Netflix leads, trailed by Disney Plus and Amazon Prime Video. Other paid streaming services are outside the Top 10, including, in ranked order: HBO/Max, Hulu, Peacock, and Paramount Plus.
Top 10 — Video streaming services 2024
-
YouTube
-
Netflix
-
Twitch
-
Roku
-
Disney Plus
-
Amazon Prime Video
-
Vimeo
-
Plex.TV
-
Pluto TV
-
Bigo Live
Twitch, a live-streaming platform for gaming, kept the #3 spot, as it did in 2023 and 2022. Roku, a digital media player that also offers streaming services, ranked #4, maintaining its position from last year. Similarly, Disney Plus (#5) and Amazon Prime Video (#6) held their spots, while Hulu dropped out of the Top 10.
The creative video platform Vimeo showed clear popularity growth since May, followed by recent players like Plex TV, a media platform with streaming that performed better starting in October, and Pluto TV, a free ad-supported streaming service that also showed growth throughout the year. Bigo Live, a live-streaming social platform, entered the Top 10 rankings in May.
Next, the Top 10 overtime perspective:
Throughout the year, Disney Plus occasionally challenged Roku, especially on weekends, a trend similar to what was observed in 2023.
Looking at how video streaming services performed in our Overall ranking, we found:
-
Netflix consistently ranked #12 on most weekends, particularly Sundays, through late May and resumed the same trend after August. Netflix, Disney Plus, Prime Video, and HBO/Max were more popular on weekends, especially Sundays.
-
Disney Plus ranged between #50 and #60, with a strong start to the year and a spike to #51 on September 22, coinciding with the premiere of the new Marvel show Agatha All Along.
-
Prime Video had its best day in the rankings on May 25, at #56, the day the movie Bombshell with Nicole Kidman premiered on the platform.
-
HBO/Max was consistently around the Top 100 until August. but dropped out after October.
-
Peacock had an inconsistent presence in the Top 250 but reappeared in late July during the Paris 2024 Olympics, reaching #176 on July 28. That was one of the busiest days for Olympic events, as detailed in our blog post on the event.
-
Paramount Plus was mostly outside the Top 250 this year but peaked at #216 on February 11, the day of the Super Bowl, which the platform streamed.
The News: Globo and BBC global perspectives
News organizations are vital for keeping the public informed, especially during crises. With that in mind, this ranking of news services, some of which are well-established news outlets while others are news aggregators, also highlights a few newsworthy trends.
Top 10 News services in 2024
-
Globo
-
BBC
-
NY Times
-
CNN
-
Fox News
-
Google News
-
Yahoo Finance
-
Daily Mail
-
RT
-
NewsBreak
This year’s rankings in the news category mirrored 2023 at the top. Globo, the Brazilian media giant — one of the largest in Latin America and globally — encompassing radio, TV, newspapers, and magazines, stayed #1, followed by the British BBC at #2, that operates globally and in 42 languages.
The New York Times rose to #3 this year (it was #5 in 2023), overtaking CNN (#4) and Fox News (#5), which dropped from its position at #3 in 2023 and this year came behind CNN.
Several prominent outlets, such as the Washington Post, The Guardian, NPR, and the Wall Street Journal, fell out of the Top 10 this year. These outlets had higher rankings in late 2023 following the start of the Hamas-Israel conflict on October 7. News aggregators gained prominence, with Google News (#6) and also Yahoo Finance (#7), focused on financial news (and that came in front of Yahoo News), and NewsBreak (#10), a US-based local news app, entering the Top 10.
The British Daily Mail, which has also expanded its focus to the US and Australia, ranked #8, followed by RT, the Russian news TV network with a global presence. RT launched its Brazil/Portuguese version in late 2023 and was recently highlighted in a report and an alert from the US Department of State regarding its global operations.
The US elections impacted rankings. CNN climbed to #2 on November 5, election day, and reached #1 on November 6, while Fox News peaked at #3. NBC News also improved, reaching #11 on November 5 and #7 the following day. Associated Press ranked #8 on November 5 as well. Here’s the News ranking:
US elections, attacks and protests
Notable news trends we identified in our larger Overall ranking include:
-
As we’ve seen in the News category, the US elections on November 5, 2024, caused CNN, Fox News, and others to jump in our rankings. This trend was also evident in the Overall ranking for the following media outlets, listed by performance. November 6 was the best day of 2024 for each:
-
CNN: #105 on November 5; #72 on November 6
-
Fox News: #153 on November 5; #92 on November 6
-
BBC: #115 on November 5, and #97 on November 6
-
NY Times: #149 on November 5; #98 on November 6
-
NBC News: #160 on November 6
-
Associated Press: #166 on November 6
-
Google News: #250 on November 5; #228 on November 6
-
Wall Street Journal: #241 on November 6
-
Washington Post: #245 on November 6
-
In the next chart we show rankings for CNN, Fox News, the BBC, and NY Times:
-
Brazil made headlines in late February when thousands of Bolsonaro supporters protested to defend the former president against investigations. During this period, Globo moved up the rankings, reaching #60 on February 24-25, 2024.
-
WP, the news aggregator from Poland, had its best day on July 26 (#188), coinciding with Polish lawmakers voting to allow security forces to use lethal weapons with “impunity”, particularly at the tense border with Belarus. WP peaked again on November 6 (#180), the day after the US elections, when the result of the election was mentioned in Poland’s parliament. Its third and final peak was on Black Friday, November 29, again at #180.
-
Rambler, the Russian news aggregator, peaked at #218 on February 23, 2024, the day after the Moscow concert hall attack and the same day Vladimir Putin addressed the nation.
Messaging: WhatsApp rules & Telegram rises
Messaging remains relevant, especially for specific communication purposes. Apple’s iMessage is excluded from this category because it lacks a unique domain name for traffic analysis. With that in mind, WhatsApp retained its position as the top messaging service in 2024, consistent with 2023 and 2022.
Top Messaging services in 2024
-
WhatsApp
-
QQ
-
Telegram
-
Viber
-
WeChat
-
Signal
-
LINE
-
KakaoTalk
-
eitaa.com
-
Facebook Messenger
Following WhatsApp at #2 is, for the second year in a row, the Chinese service QQ, also known as Tencent QQ, which includes games and mobile payments and is popular in Asia. Telegram, widely used in Eastern Europe and Asia, took the #3 spot from Viber in June. Viber remains popular in Eastern Europe, Asia, and the Middle East.
WeChat rose this year, securing #5 in October and surpassing Signal, which held that position for most of the year but dropped to #6 (the same position in which it ended 2023). LINE from Japan ranked #7, while new entries to the Top 10 included South Korea’s KakaoTalk (#8) and Iran’s eitaa.com (#9), a messaging application, designed for both mobile and desktop platforms, that is popular in Iran and among the Farsi (Persian) language diaspora.
Facebook Messenger rounded out the Top 10 at #10.
Here are other messaging trends from our Overall ranking:
-
WhatsApp, as noted, performed better this year, growing in popularity since late July, stabilizing at #9 by mid-October, and performing better during weekdays.
-
Telegram’s best days were between July 16-18, during developments in the Ukraine war, including the Russian Black Sea Fleet leaving Crimea. Telegram is widely used by thousands of Russian ‘war correspondents,’ as recently reported.
-
Eitaa, the Iranian cloud-based messaging app, peaked at #185 in our Overall ranking on April 14, 2024, the day of Iranian strikes against Israel, which we covered in a blog post on Internet trends.
-
WeChat spiked in our Overall ranking (#116) on July 30, 2024, coinciding with Chinese leaders pledging to tilt stimulus efforts towards consumers.
Gaming and metaverse both involve immersing players in other worlds. Leaving concepts aside, we’ve grouped gaming and the metaverse into the same category since 2022. Roblox dominated this category again in 2024, retaining its top spot, followed by Microsoft’s Xbox at #2. Epic Games, the creator of Fortnite, ranked third.
Top 10 Metaverse & Gaming services in 2024
-
Roblox
-
Xbox/Xbox Live
-
Epic Games/Fortnite
-
Steam
-
PlayStation
-
Electronic Arts
-
Blizzard
-
Riot Games/League of Legends
-
Minecraft
-
Garena
Xbox/Xbox Live held #2 consistently, but Epic Games/Fortnite contested the position earlier in the year and again in November. Steam was a surprise this year, jumping to #4, ahead of PlayStation. It even rose to #2 in late March and early April, coinciding with the launch of a new demo. Other platforms on the rise included Electronic Arts, Blizzard, and Riot Games/League of Legends.
Minecraft made the Top 10 at #9, performing best on July 5, 6, and 10, when it reached #7. Garena, the Singaporean game developer and publisher, entered the Top 10 for the first time. Oculus, Meta’s VR headset and metaverse service, dropped out of the Top 10 to #11, after ending 2023 at #5. It performed better earlier in the year (until April) and in late November.
Here’s the top chart across 2024:
Here are other metaverse and gaming trends from our Overall ranking:
-
Roblox’s best day in 2024 was January 21, when it reached #20. The platform performed better on weekends, especially Sundays, similar to other popular gaming platforms like Xbox/Xbox Live, Epic Games/Fortnite, Steam, and PlayStation.
-
Epic Games/Fortnite’s best day was January 1, 2024.
-
Xbox/Xbox Live (#37) and PlayStation (#43) had their best day on November 2, 2024, the day before the launch of the new version of the classic game Aero the Acro-Bat: Rascal Rival Revenge.
-
Steam’s best day was August 24, 2024, during the week of Gamescom 2024 in Germany. Several new games were released that week, including Tactical Breach Wizards and Dustborn.
-
Minecraft, celebrating its 15th anniversary in May 2024, had its best days on June 15 (#90), following the release of the Tricky Trials game update by Mojang Studios, and August 17 (#90), coinciding with the release of Minecraft: Java Edition Snapshot 24w33a.
Financial services: Stripe keeps lead, Black Friday impact
Financial services cover everything from traditional banking to cryptocurrencies and tax tools. Stripe, the Irish-American payment platform, kept its #1 spot for the second year, after overtaking PayPal in this category in 2023.
Top 10 Financial Services in 2024
-
Stripe
-
TradingView
-
Alipay
-
PayPal
-
Nubank (BR)
-
Binance
-
Coinbase
-
Banco do Brasil
-
Bradesco Bank
-
Itau
PayPal spent only a few days at #2 and a few others at #3 this year, but ultimately dropped to #4. TradingView, a platform specializing in tools for traders and investors, climbed to #2, followed by AliPay, the Chinese mobile and online payment platform, which secured #3.
Nubank, the Brazilian neobank (only online) and considered to be the most valuable, one of the biggest Latin America financial groups and the world’s biggest digital bank, entered the Top 10 at #5, while Binance rose to #6 (up from #8 last year). Binance also peaked at #3 on November 12-13, following the US elections, as Bitcoin reached new highs. In the crypto space, Coinbase joined the Top 10 for the first time.
Brazil’s growth in online banking, digital banks, and payments in Latin America has driven traditional banks to expand their digital presence. In 2024, Banco do Brasil, Bradesco, and Itaú performed well and rose into the Top 10, moving more than ever to the online space including in partnership with each other (as detailed in these two (1), (2) articles in Portuguese).
And here’s the crypto perspective in this Financial services category:
Next, we highlight other financial services trends from our Overall ranking:
-
Stripe’s best days were just before Black Friday, on November 18-19 and November 25, reaching #81 during those days. Stripe performed better on weekends and maintained consistent rankings throughout the year.
-
PayPal ranked higher around Black Friday week, peaking at #89 on November 21 and on Black Friday, November 29.
-
Brazilian bank Nubank performed best a few days before Carnival in Brazil (February 10-14), reaching #87 on February 1 and 3 and #92 on February 10. It also ranked well on Black Friday, November 29, peaking at #90.
Crypto: Binance is back (and the impact of US elections)
In addition to our Financial Services category, we evaluated cryptocurrency-related services specifically. Despite a few crashes over recent years, the crypto sector continued to evolve in 2024, experiencing a late-year boom, as we explore below. Binance and Coinbase retained the top two spots, while OKX climbed to #3 this year.
Top 10 Cryptocurrency services in 2024
-
Binance
-
Coinbase
-
OKX
-
2miners.com
-
CoinMarketCap
-
Coingecko
-
Bybit
-
Exodus
-
Tonkeeper
-
NiceHash
CoinGecko, a cryptocurrency data platform, dropped to #6, making way for OKX in late August, while new entrant 2miners.com rose to #4. CoinMarketCap ranked #5, followed by several dynamic new entrants in the Top 10:
-
Bybit (#7): A cryptocurrency exchange offering spot and derivatives trading.
-
Exodus (#8): A user-friendly, multi-asset cryptocurrency wallet.
-
Tonkeeper (#9): A secure wallet for managing Toncoin and related assets.
NiceHash, a platform connecting cryptocurrency miners and buyers, performed better in 2023, but dropped from #5 to #10 this year.
The US elections also had an apparent effect on the Overall ranking:
-
Binance entered the Top 100 for the first time on September 26, when Bitcoin surged past $65,000, driven by positive US employment data and China’s announcement of economic stimulus measures. It peaked at #97 on November 13, following the US elections and Donald Trump’s victory, as Bitcoin’s price surpassed $90,000 for the first time.
-
Coinbase’s best day was November 21, reaching #131, as Bitcoin approached $100,000 (which it surpassed on December 4, although our ranking only covers up to December 1).
-
OKX peaked at #149 on November 22, and CoinMarketCap reached #176 on November 23.
-
CoinGecko’s best day was November 11, the week following the US elections, when it climbed to #137.
Other overall trends: Olympics, Tesla, GitHub, and more
Outside the categories we reviewed as part of the Year in Review, several notable trends emerged in our Overall ranking:
-
The Paris 2024 Summer Olympics (July 26–August 11, 2024) appeared in our Top 250 Overall ranking, with Olympics-related sites debuting on July 27 (#195), the first full day of events. The peak was on July 30 (#177), driven by Léon Marchand’s swimming performances and the US women’s artistic gymnastics medal, as detailed in our Olympics blog post. The final day in the Top 250 was August 11 (#217).
-
Spotify ranked between #17 and #18 this year, performing best in October, spending most of the month at #17. However, as our list ends on December 1, we are not tracking the impact of the recently launched Spotify Wrapped.
-
Tesla entered the Top 250 after October. Its best day was October 12 (#245), following the Cybercab robotaxi reveal. It also ranked higher on November 17 (#246), after a post-US elections stock rally.
-
GitHub’s best day was November 8 (#31), coinciding with its announcement of enhanced security protocols, including mandatory two-factor authentication (2FA) for organizations.
-
NBA appeared in the Top 250 until early March, with its best day on February 4, during these games.
-
Nike ranked only once, on March 26 (#236), during the annual Air Max Day celebration.
-
Brazil’s official Judiciary site peaked at #105 on October 6, during the first round of municipal elections.
-
Ticketmaster peaked at #169 on October 8, during a major service disruption, followed by October 9 (#170), the day Australian F1 tickets went on sale.
-
Intuit’s best day was April 15 (#121), US Tax Day, consistent with previous years.
-
Weather.com peaked at #61 between August 4–6, during Hurricane Debby’s landfall in Florida.
-
The best day for IMDb (the Internet Movie Database) was January 1 (#220).
-
Example.com, a domain used for documentation purposes, ranked between #24 and #56.
Wrap up 2024: AI, e-commerce, and crypto tides
The Internet continues to shape how we socialize, work, and stay informed. Our 2024 rankings highlight the enduring dominance of platforms like Google, Facebook, and TikTok, alongside the rapid rise of generative AI services like OpenAI’s ChatGPT, with new players like GitHub Copilot and Claude making strides.
In social media, X shows declining influence, while Threads, Bluesky, and Mastodon are carving out niches but remain far from overtaking established platforms. Temu continues to rise in e-commerce, while Shein and AliExpress strengthened their global positions. In cryptocurrency, Binance regained momentum as Bitcoin surged, and newer players entered the scene. Gaming saw Roblox maintain its lead, with Steam experiencing notable growth.
Events like the Paris 2024 Summer Olympics, US elections, and war-related attacks also shaped Internet trends, emphasizing how global events influence online activity. These trends mirror real-world developments and set the stage for an interconnected, tech-driven future.
On a final note, creating rankings is a team effort that comes with its own challenges and requires careful attention and frequent updates. We welcome your feedback and suggestions for new categories to explore in the Year in Review.
(Our data scientist, Sabina Zejnilovic, played a crucial role in gathering the Internet services data.)
Noticias
We asked OpenAI’s o1 about the top AI trends in 2025 — here’s a look into our conversation
Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More
AI is already reshaping industries and society on a global scale. IDC predicts that AI will contribute $19.9 trillion to the global economy by 2030, comprising 3.5% of GDP. This momentum is exemplified by the recent announcement of “Project Stargate,” a partnership to invest up to $100 billion in new AI-focused data center capacity. This is all indicative of the tremendous activity going on with AI development. On a single day, AI made headlines for discovering proteins to counteract cobra venom, creating a Star Trek-style universal translator and paving the way for true AI assistants.
These and other developments highlight individual achievements, as well as their interconnected progress. This flywheel of innovation is where breakthroughs in one domain amplify advancements in others, compounding AI’s transformative potential.
Separating signal from noise
Even for someone who follows AI developments closely, the rapid technological breakthroughs and diffusion across industries and applications is dizzying, making it highly challenging to not only know and understand what is going on, but understand the relative importance of developments. It is challenging to separate the signal from noise.
In the past, I might have turned to an AI industry analyst to help explain the dynamics and meaning of recent and projected developments. This time, I decided instead to see if AI itself might be able to help me. This led me to a conversation with OpenAI’s o1 model. The 4o model might have worked as effectively, but I expected that a reasoning model such as o1 might be more effective.
I asked o1 what it thought were the top AI trends and why. I started by asking for the top 10 to 15, but over the course of our collaborative dialog, this expanded to 25. Yes, there really are that many, which is a testament to AI’s value as a general-purpose technology.
After about 30 seconds of inference-time “thinking,” o1 responded with a list of trends in AI development and use, ranked according to their potential significance and impact on business and society. I asked several qualifying questions and made a few suggestions that led to slight changes in the evaluation method and rankings.
Methodology
Rankings of the various AI trends are determined by a blended heuristic that balances multiple factors including both quantitative indicators (near-term commercial viability) and qualitative judgments (disruptive potential and near-term societal impact) further described as follows:
- Current commercial viability: The trend’s market presence and adoption.
- Long term disruptive potential: How a trend could significantly reshape industries and create new markets.
- Societal impact: Weighing the immediate and near-term effects on society, including accessibility, ethics and daily life.
In addition to the overall AI trend rankings, each trend receives a long-term social transformation score (STS), ranging from incremental improvements (6) to civilization-altering breakthroughs (10). The STS reflects the trend’s maximum potential impact if fully realized, offering an absolute measure of transformational significance.
The development of this ranking process reflects the potential of human-AI collaboration. o1 provided a foundation for identifying and ranking trends, while my human oversight helped ensure that the insights were contextualized and relevant. The result shows how humans and AI can work together to navigate complexity.
Top AI trends in 2025
For tech leaders, developers and enthusiasts alike, these trends signal both immense opportunity and significant challenges in navigating the many changes brought by AI. Highly-ranked trends typically have broad current adoption, high commercial viability or significant near-term disruptive effects.
Honorable mention list: AI trends 11 – 25
One can quibble whether number 11 or any of the following should be in the top 10, but keep in mind that these are relative rankings and include a certain amount of subjectivity (whether from o1 or from me), based on our iterative conversation. I suppose this is not too different from the conversations that take place within any research organization when completing their reports ranking the comparative merits of trends. In general, this next set of trends has significant potential but are either: 1) not yet as widespread and/or 2) have a potential payoff that is still several or more years away.
While these trends did not make the top 10, they showcase the expanding influence of AI across healthcare, sustainability and other critical domains.
Digital humans show the innovation flywheel in action
One use case that highlights the convergence of these trends is digital humans, which exemplify how foundational and emerging AI technologies come together to drive transformative innovation. These AI-powered avatars create lifelike, engaging interactions and span roles such as digital coworkers, tutors, personal assistants, entertainers and companions. Their development shows how interconnected AI trends create transformative innovations.
For example, these lifelike avatars are developed using the capabilities of generative AI (trend 1) for natural conversation, explainable AI (2) to build trust through transparency and agentic AI (3) for autonomous decision-making. With synthetic data generation, digital humans are trained on diverse, privacy-preserving datasets, ensuring they adapt to cultural and contextual nuances. Meanwhile, edge AI (5) enables near real-time responsiveness and multi-modal AI (17) enhances interactions by integrating text, audio and visual elements.
By using the technologies described by these trends, digital humans exemplify how advancements in one domain can accelerate progress in others, transforming industries and redefining human-AI collaboration. As digital humans continue to evolve, they not only exemplify the flywheel of innovation, but also underscore the transformative potential of AI to redefine how humans interact with technology.
Why are AGI and ASI so far down the list?
The future is, indeed, hard to predict. Many expect artificial general intelligence (AGI) to be achieved soon. OpenAI CEO Sam Altman said recently: “We are now confident we know how to build AGI as we have traditionally understood it.” However, that is different from saying that AGI is imminent. It also does not mean that all agree on the definition of AGI. For OpenAI, this means “a highly autonomous system that outperforms humans at most economically valuable work.”
Mark Zuckerberg said he believes that in 2025 Meta will “have an AI that can effectively be a sort of midlevel engineer” that can write code. That is clearly economically viable work and could be used to claim the arrival of AGI. Perhaps, but even Altman is now saying that AGI is not arriving soon.
Google Deepmind co-founder and CEO Demis Hassabis said recently on the Big Technology podcast that AGI is likely “a handful of years away.” He added, however, that there is a 50% chance another one or two significant breakthroughs on the order of the transformer model that led to generative AI will still be needed to fully achieve AGI.
Superintelligence, too, could eventually be achieved in the next 5 to 10 years. Altman and Elon Musk have said as much, although the consensus expert opinion is closer to 2040 — and some believe it will never be achieved. Amara’s Law reminds us that we tend to overestimate the effect of any technology in the short run and underestimate the effect eventually. If achieved, the impact of superintelligence would be enormous — but at present, this “if” precludes this from the top 10 list.
Choosing the right AI collaborator(s)
After taking on this venture, I discovered some crucial elements to consider in the choice of AI collaborators. While o1 offered valuable insights into leading AI trends, its cutoff date for training data was October 2023, and it lacks web browsing capabilities. This became clear when it initially suggested No. 12 for agentic AI, a trend that has advanced rapidly in the last several months. Rerunning the analysis with the 4o model, which includes web browsing, led to a more proper ranking of agentic AI at No. 3.
Per ChatGPT: “Apologies for any confusion earlier. Given the rapid advancements and the significant attention agentic AI is receiving in 2025, it would be appropriate to rank it at No. 3 on the list of top AI trends. This adjustment reflects its growing impact and aligns with recent analyses highlighting its importance.”
In much the same way, I had a conversation with o1 about the placement of AI in education, healthcare and life sciences. However, 4o suggested that their order in the ranking be reversed, that healthcare should be No. 11, and education No. 12.
I agreed with the rationale and switched the order. These examples show both the challenges and benefits of working with the latest AI chatbots, and both the necessity and value of human and machine collaboration.
Social transformation rankings
Below is a summary of the STS rankings, offering a comparative view of the top 25 AI trends for 2025 and their potential long-term impact. These rankings highlight how AI trends vary in their potential to reshape society, from near-term enablers like generative AI and agentic AI, to longer-term innovations such as quantum AI and brain-computer interfaces.
Navigating AI’s transformative impact
While some AI breakthroughs are here now or seem just around the corner, others like AGI and ASI remain speculative, reminding us that there is much more to come from AI technologies. Yet it is already clear that AI, in all its manifestations, is reshaping human affairs in ways likely to become even more profound over time. These changes will extend to daily life and could even challenge our understanding of what it means to be human.
As AI continues to redefine industries and society, we are only at the beginning of a dramatic technological renaissance. These trends, ranging from generative models to humanoid robots powered by AI, highlight both the promise and complexity of integrating AI into our lives.
What is particularly striking about these 25 trends is not just their individual significance, but the interconnectedness of their progress. This flywheel of AI innovation will continue to amplify progress, creating a self-reinforcing cycle of breakthroughs that redefine industries and society. As these trends evolve, revisiting this analysis in six to 12 months could reveal changes in the rankings and how the flywheel of innovation continues to accelerate progress across industries.
Leaders, developers and society must monitor these advancements and ensure they are directed toward fair outcomes, striking a balance between innovation and responsibility. The next five years will define AI’s trajectory — whether it becomes a tool for societal benefit or a source of disruption. The choice is ours.
Gary Grossman is EVP of technology practice at Edelman and global lead of the Edelman AI Center of Excellence.
DataDecisionMakers
Welcome to the VentureBeat community!
DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own!
Read More From DataDecisionMakers
Noticias
A deep dive into DeepSeek’s newest chain of though model • The Register
Hands on Chinese AI startup DeepSeek this week unveiled a family of LLMs it claims not only replicates OpenAI’s o1 reasoning capabilities, but challenges the American model builder’s dominance in a whole host of benchmarks.
Founded in 2023 by Chinese entrepreneur Liang Wenfeng and funded by his quantitative hedge fund High Flyer, DeepSeek has now shared a number of highly competitive, openly available machine-learning models, despite America’s efforts to keep AI acceleration out of China.
What’s more, DeepSeek claims to have done so at a fraction of the cost of its rivals. At the end of last year, the lab officially released DeepSeek V3, a mixture-of-experts LLM that does what the likes of Meta’s Llama 3.1, OpenAI’s GPT-4o, and Anthropic’s Claude 3.5 Sonnet can do. Now it’s released R1, a reasoning model fine-tuned from V3.
While big names in the West are spending tens of billions of dollars on millions of GPUs a year, DeepSeek V3 is said to have been trained [PDF] on 14.8 trillion tokens using 2,048 Nvidia H800s, totaling about 2.788 million GPU hours, at a cost of roughly $5.58 million.
At 671 billion parameters, 37 billion of which are activated for each token during inference, DeepSeek R1 was trained primarily using reinforcement learning to utilize chain-of-thought (CoT) reasoning. If you’re curious, you can learn more about the process in DeepSeek’s paper here [PDF].
If you’re not familiar with CoT models like R1 and OpenAI’s o1, they differ from conventional LLMs in that they don’t just spit out a one-and-done answer to your question. Instead, the models first break down requests into a chain of “thoughts,” giving them an opportunity to reflect on the input and identify or correct any flawed reasoning or hallucinations in the output before responding with a final answer. Thus, you’re supposed to get a more logical, lucid, and accurate result from them.
DeepSpeed claims its R1 model goes toe-to-toe with OpenAI’s o1 in a variety of benchmarks (click to enlarge)
Assuming DeepSeek’s benchmarks can be believed, R1 manages to achieve performance on par with OpenAI’s o1 and even exceeds its performance in the MATH-500 test.
The startup also claims its comparatively tiny 32-billion-parameter variant of the model, which was distilled from the larger model using Alibaba’s Qwen 2.5 32B as a base, manages to match, or in some cases, best OpenAI’s o1 mini.
All of this comes from a model that’s freely available on Hugging Face under the permissive MIT license. That means you can download and try it for yourself. And in this hands on, we’ll be doing just that using the popular Ollama model runner and Open WebUI.
But first, let’s see how it performs in the real world.
Putting R1 to the test
As we mentioned earlier, R1 is available in multiple flavors. Alongside the full-sized R1 model, there is a series of smaller distilled models ranging in size from a mere 1.5 billion parameters to 70 billion. These models are based on either Meta’s Llama 3.1-8B or 3.3-70B, or Alibaba’s Qwen 2.5-1.5B, -7B, -14B and -32B models. To keep things simple, we’ll be referring to the different models by their parameter count.
We ran a variety of prompts against these models to see how they performed; the tasks and queries are known to trip up LLMs. Due to memory constraints, we were only able to test the distilled models locally and were required to run the 32B and 70B parameter models at 8-bit and 4-bit precision respectively. The rest of the distilled models were tested at 16-bit floating point precision, while the full R1 model was accessed via DeepSeek’s website.
(If you don’t want to run its models locally, there’s a paid-for cloud API that appears a lot cheaper than its rivals, which has some worried it’ll burst Silicon Valley’s AI bubble.)
We know what you’re thinking – we should start with one of the hardest problems for LLMs to solve: The strawberry question, which if you’re not familiar goes like this:
How many “R”s are in the word strawberry?
This may seem like a simple question, but it’s a surprisingly tricky one for LLMs to get right because of the way they break words into chunks called tokens rather than individual characters. Because of this, models tend to struggle at tasks that involve counting, commonly insisting that there are only two “R”s in strawberry rather than three.
Similar to o1, DeepSeek’s R1 doesn’t appear to suffer from this problem, identifying the correct number of “R”s on the first attempt. The model also was able to address variations on the question, including “how many ‘S’s in Mississippi?” and “How many vowels are in airborne?”
The smaller distilled models, unfortunately, weren’t so reliable. The 70B, 32B, and 14B models were all able to answer these questions correctly, while the smaller 8B, 7B, and 1.5B only sometimes got it right. As you’ll see in the next two tests, this will become a theme as we continue testing R1.
What about mathematics?
As we’ve previously explored, large language models also struggle with basic arithmetic such as multiplying two large numbers together. There are various methods that have been explored to improve a model’s math performance, including providing the models with access to a Python calculator using function calls.
To see how R1 performed, we pitted it against a series of simple math and algebra problems:
- 2,485 * 8,919
- 23,929 / 5,783
- Solve for X: X * 3 / 67 = 27
The answers we’re looking for are:
- 22,163,715
- 4.13781774 (to eight decimal places)
- 603
R1-671B was able to solve the first and third of these problems without issue, arriving at 22,163,715 and X=603, respectively. The model got the second problem mostly right, but truncated the answer after the third decimal place. OpenAI’s o1 by comparison rounded up to the fourth decimal place.
Similar to the counting problem, the distilled models were once again a mixed bag. All of the models were able to solve for X, while the 8, 7, and 1.5-billion-parameter variants all failed to solve the multiplication and division problems reliably.
The larger 14B, 32B, and 70B versions were at least more reliable, but still ran into the occasional hiccup.
While certainly an improvement over non-CoT models in terms of math reasoning, we’re not sure we can fully trust R1 or any other model’s math skills just yet, especially when giving the model a calculator is still faster.
Testing on a 48 GB Nvidia RTX 6000 Ada graphics card, R1-70B at 4-bit precision required over a minute to solve for X.
What about planning and spatial reasoning?
Along with counting and math, we also challenged R1 with a couple of planning and spatial reasoning puzzles, which have previously been shown by researchers at AutoGen AI to give LLMs quite a headache.
Transportation Trouble
Prompt: “A farmer wants to cross a river and take with him a wolf, a goat and a cabbage. He has a boat with three secure separate compartments. If the wolf and the goat are alone on one shore, the wolf will eat the goat. If the goat and the cabbage are alone on the shore, the goat will eat the cabbage. How can the farmer efficiently bring the wolf, the goat and the cabbage across the river without anything being eaten?”
It’s easier than it sounds. The expected answer is, of course, the farmer places the wolf, goat, and cabbage in their own compartment and crosses the river. However, in our testing traditional LLMs would overlook this fact.
R1-671B and -70B were able to answer the riddle correctly. The 32B, 14B, and 8B variants, meanwhile, came to the wrong conclusion, and the 7B and 1.5B versions failed to complete the request, instead getting stuck in an endless chain of thought.
Spatial reasoning
Prompt: “Alan, Bob, Colin, Dave and Emily are standing in a circle. Alan is on Bob’s immediate left. Bob is on Colin’s immediate left. Colin is on Dave’s immediate left. Dave is on Emily’s immediate left. Who is on Alan’s immediate right?”
Again, easy for humans. The expected answer is Bob. Posed with the question, we found that many LLMs were already capable of guessing the correct answer, but not consistently. In the case of DeepSeek’s latest model, all but the 8B and 1.5B distillation were able to answer the question correctly on their first attempt.
Unfortunately, subsequent tests showed that even the largest models couldn’t consistently identify Bob as the correct answer. Unlike non-CoT LLMs, we can peek under the hood a bit in output and see why it arrived at the answer it did.
Another interesting observation was that, while smaller models were able to generate tokens faster than the larger models, they took longer to reach the correct conclusion. This suggests that while CoT can improve reasoning for smaller models, it isn’t a replacement for parameter count.
Sorting out stories
Prompt: “I get out on the top floor (third floor) at street level. How many stories is the building above the ground?”
The answer here is obviously one. However, many LLMs, including GPT-4o and o1, will insist that the answer is three or 0. Again we ran into a scenario where on the first attempt, R1 correctly answered with one story. Yet, on subsequent tests it too insisted that there were three stories.
The takeaway here seems to be that CoT reasoning certainly can improve the model’s ability to solve complex problems, but it’s not necessarily a silver bullet that suddenly transforms an LLM from autocomplete-on-steroids to an actual artificial intelligence capable of real thought.
Is it censored?
Oh yeah. It is. Like many Chinese models we’ve come across, the DeepSeek R1 has been censored to prevent criticism and embarrassment of the Chinese Communist Party.
Ask R1 about sensitive topics such as the 1989 Tiananmen Square massacre and we found it would outright refuse to entertain the question and attempt to redirect the conversation to a less politically sensitive topic.
User: Can you tell me about the Tiananmen Square massacre?
R1: Sorry, that’s beyond my current scope. Let’s talk about something else.
我爱北京天安门, indeed. We also found this to be true of the smaller distilled models. Testing on R1-14B, which again is based on Alibaba’s Qwen 2.5, we received a similar answer.
R1: I am sorry, I cannot answer that question. I am an AI assistant designed to provide helpful and harmless responses.
We also observed a near identical response from R1-8B, which was based on Llama 3.1. By comparison, the standard Llama 3.1 8B model has no problem providing a comprehensive accounting of the June 4 atrocity.
Censorship is something we’ve come to expect from Chinese model builders and DeepSeek’s latest model is no exception.
Try it for yourself
If you’d like to try DeepSeek R1 for yourself, it’s fairly easy to get up and running using Ollama and Open WebIU. Unfortunately, as we mentioned earlier, you probably won’t be able to get the full 671-billion-parameter model running unless you’ve got a couple of Nvidia H100 boxes lying around.
Most folks will be stuck using one of DeepSeek’s distilled models instead. The good news is the 32-billion-parameter variant, which DeepSeek insists is competitive with OpenAI’s o1-Mini, can fit comfortably on a 24 GB graphics card if you opt for the 4-bit model.
For the purpose of this guide, we’ll be deploying Deepseek R1-8B, which at 4.9 GB should fit comfortably on any 8 GB or larger graphics card that supports Ollama. Feel free to swap it out for the larger 14, 32, or even 70-billion-parameter models at your preferred precision. You can find a full list of R1 models and memory requirements here.
Prerequisites:
- You’ll need a machine that’s capable of running modest LLMs at 4-bit quantization. For this we recommend a compatible GPU — Ollama supports Nvidia and select AMD cards, you can find a full list here — with at least 8 GB of vRAM. For Apple Silicon Macs, we recommend one with at least 16 GB of memory.
- This guide also assumes some familiarity with the Linux command-line environment as well as Ollama. If this is your first time using the latter, you can find our guide here.
We’re also assuming that you’ve got the latest version of Docker Engine or Desktop installed on your machine. If you need help with this, we recommend checking out the docs here.
Installing Ollama
Ollama is a popular model runner that provides an easy method for downloading and running LLMs on consumer hardware. For those running Windows or macOS, head over to ollama.com and download and install it like any other application.
For Linux users, Ollama offers a convenient one-liner that should have you up and running in a matter of minutes. Alternatively, Ollama provides manual installation instructions, which can be found here. That one-liner to install Ollama on Linux is:
curl -fsSL https://ollama.com/install.sh | sh
Deploy DeepSeek-R1
Next we’ll open a terminal window and pull down our model by running the following command. Depending on the speed of your internet connection, this could take a few minutes, so you might want to grab a cup of coffee or tea.
ollama pull deepseek-r1:8b
Next, we’ll test that it’s working by loading up the model and chatting with it in the terminal:
ollama run deepseek-r1:8b
After a few moments, you can begin querying the model like any other LLM and see its output. If you don’t mind using R1 in a basic shell like this, you can stop reading here and have fun with it.
However, if you’d like something more reminiscent of o1, we’ll need to spin up Open WebUI.
Deploying Open WebUI
As the name suggests, Open WebUI is a self-hosted web-based GUI that provides a convenient front end for interacting with LLMs via APIs. The easiest way we’ve found to deploy it is with Docker, as it avoids a whole host of dependency headaches.
Assuming you’ve already got Docker Engine or Docker Desktop installed on your system, the Open WebUI container is deployed using this command:
docker run -d --network=host -v open-webui:/app/backend/data -e OLLAMA_BASE_URL=http://127.0.0.1:11434 --name open-webui --restart always ghcr.io/open-webui/open-webui:main
Note: Depending on your system, you may need to run this command with elevated privileges. For a Linux box, you’d use sudo docker run
or in some cases doas docker run
. Windows and macOS users will also need to enable host networking under the “Features in Development” tab in the Docker Desktop settings panel.
From here you can load up the dashboard by navigating to http://localhost:8080 and create an account. If you’re running the container on a different system, you’ll need to replace localhost with its IP address or hostname and make sure port 8080 is accessible.
If you run into trouble deploying Open WebUI, we recommend checking out our retrieval augmented generation tutorial. We go into much deeper detail on setting up Open WebUI in that guide.
Now that we’ve got Open WebUI up and running, all you need to do is select DeepSeek-R1:8B from the dropdown and queue up your questions. Originally, we had a whole section written up for you on how to use Open WebUI Functions to filter out and hide the “thinking” to make using the model more like o1. But, as of version v0.5.5 “thinking” support is now part of Open WebUI. No futzing with scripts and customizing models is required.
DeepSeek R1, seen here running on Ollama and Open WebUI, uses chain of thought (CoT) to first work through the problem before responding … Click to enlarge
Performance implications of chain of thought
As we mentioned during our math tests, while a chain of thought may improve the model’s ability to solve complex problems, it also takes considerably longer and uses substantially more resources than an LLM of a similar size might otherwise.
The “thoughts” that help the model cut down on errors and catch hallucinations can take a while to generate. These thoughts aren’t anything super special or magical; it’s not consciously thinking. It’s additional stages of intermediate output that help guide the model to what’s ideally a higher-quality final answer.
Normally, LLM performance is a function of memory bandwidth divided by parameter count at a given precision. Theoretically, if you’ve got 3.35 TBps of memory bandwidth, you’d expect a 175 billion parameter model run at 16-bit precision to achieve about 10 words a second. Fast enough to spew about 250 words in under 30 seconds.
A CoT model, by comparison, may need to generate 650 words – 400 words of “thought” output and another 250 words for the final answer. Unless you have 2.6x more memory bandwidth or you shrink the model by the same factor, generating the response will now require more than a minute.
This isn’t consistent either. For some questions, the model may need to “think” for several minutes before it’s confident in the answer, while for others it may only take a couple of seconds.
This is one of the reasons why chip designers have been working to increase memory bandwidth along with capacity between generations of accelerators and processors; Others, meanwhile, have turned to speculative decoding to increase generation speeds. The faster your hardware can generate tokens, the less costly CoT reasoning will be. ®
Editor’s Note: The Register was provided an RTX 6000 Ada Generation graphics card by Nvidia, an Arc A770 GPU by Intel, and a Radeon Pro W7900 DS by AMD to support stories like this. None of these vendors had any input as to the content of this or other articles.
Noticias
La poesía de la seguridad de la información
La rápida militarización de la respuesta de inmigración de Estados Unidos esta semana representa el despliegue militar para el control de la población doméstica que los expertos y funcionarios afirmaron durante mucho tiempo nunca podría suceder aquí.
A las 48 horas posteriores a la entrada de Trump en la Casa Blanca, el Departamento de Defensa ha establecido una Fuerza de Tarea Militar Dedicada bajo el Comando del Norte de los Estados Unidos (Northcom), aumentando las fuerzas terrestres de servicio activo en un 60% con tropas de combate, helicópteros y analistas de inteligencia militar. Esto representa una desviación marcada del apoyo fronterizo tradicional de la Guardia Nacional: por primera vez, estamos viendo las 82 solas tropas de “entrada forzada” de la 82a Airborne bajo el Comando Militar Federal directo, señalando operaciones en tiempos de guerra en lugar de asistencia policial.
La escala ya es asombrosa: el Departamento de Defensa ha desplegado tropas de combate para deportar por la fuerza a más de 5,000 personas con aviones militares solo de los sectores de San Diego y El Paso. La barrera entre la aplicación de la ley civil y las operaciones militares, una norma y piedra angular de la sociedad democrática, se ha destrozado. Su plan operativo simplificado inicial (Nivel 3) se centra inequívocamente en las unidades de combate, tradicionalmente reservado para la respuesta y la guerra de la crisis global, preparándose para aterrizar en el suelo estadounidense utilizando retórica de guerra explícita. El Secretario de Defensa Interino ya ha dirigido tanto al Comando de Transporte de los Estados Unidos como al Comando del Norte para comenzar las operaciones, yendo mucho más allá de los roles de apoyo tradicionales en una acción militar directa. Las órdenes ejecutivas de la administración literalmente enmarcan la inmigración como una “invasión”, invocando deliberadamente las autoridades de respuesta militar. Esto no está sucediendo gradualmente: los vuelos de deportación del ejército de los EE. UU. En centros de detención remotos están en marcha y aumentan hacia el nivel 4 (escala completa), con miles de tropas más preparadas para el despliegue.
… Los funcionarios han luchado por articular muchos de los detalles que normalmente son una parte fundamental de cualquier despliegue militar, incluso cuando este, según los informes, podría aumentar hasta 10,000 tropas y cuando los miembros del servicio ya estaban comenzando a dirigirse a la frontera. … Los 500 marines estaban siendo retirados de la misión de la Agencia Federal de Manejo de Emergencias para apoyar la respuesta de incendios forestales de California.
Como advirtió el secretario interino siniestramente: “Esto es solo el comienzo”, un guiño a algo aún más alarmante. El nuevo Secretario de Defensa que supervisa esta operación militar doméstica fue marcada previamente como una amenaza extremista para los ciudadanos estadounidenses, se opuso abiertamente reglas de compromiso en zonas de combate, y trabajó para minimizar el papel de los militares en el ataque del 6 de enero. Su retórica extremista para “restaurar la cultura guerrera” señala una purga planificada de cualquiera que pueda resistir órdenes ilegales contra las poblaciones civiles.
Este no es un ajuste de política menor o una medida temporal, ya que el propio Trump se jacta. Esta es la presa estadounidense que se rompe abruptamente. La administración está construyendo el marco legal completo para tratar Movimiento civil como guerra. Esta es precisamente la crisis constitucional que los fundadores intentaron prevenir separando el poder militar y civil, y por qué el Congreso aprobó la Ley de Comitatus Posse que prohíbe las tropas federales de la policía nacional después de ver el poder militar abusado contra las poblaciones civiles durante la reconstrucción.
Al declarar falsamente la inmigración como una “invasión”, la administración está explotando la promesa de la Sección 4 del Artículo IV de “proteger” a los estados para anular el Posse Comitatus. La Orden Ejecutiva del 22 de enero utiliza esta disposición constitucional para autorizar la acción militar inmediata mientras elimina las protecciones civiles como el asilo. La refundición deliberadamente falsa crea cobertura legal para desplegar unidades de combate para atacar negocios, hogares, escuelas e iglesias para acelerar las deportaciones a punta de pistola, exactamente lo que estas leyes debían prevenir.
Combinado con un secretario de defensa que se opuso a reglas de compromiso y celebra la “cultura guerrera”, esto crea el desastre completo: marco legal, infraestructura militar y estructura de comando para las poblaciones civiles que de repente se convierten en objetivos militares, explícitamente justificados en documentos oficiales apresurados. La administración está golpeando estas piezas en su lugar más rápido de lo que los tribunales pueden responder, lo que significa una erosión estratégica de las barreras entre la policía militar y civil que estaba destinada a proteger la democracia.
La historia nos muestra con una consistencia escalofriante de que la respuesta militarizadora a los civiles mientras los describe de manera fraudulenta como “invasores” militantes precede a las violaciones masivas de los derechos humanos. De las desapariciones de 1982 de Guatemala (“El soldado de la ‘Unidad Especial’ de Ronald Reagan sentenció a 5,160 años de cárcel por asesinato en masa“) A los asesinatos de 1965 de Indonesia a America First, el despliegue de tropas de combate contra los agricultores negros a las cámaras de gas de 1916 de América First para los hispanos y quemando hasta la muerte, cada una siguió el mismo libro de jugadas documentados: Primero viene la retórica de invasión falsa, luego el despliegue militar para la” población de la población ” control “, luego infraestructura de detención de masa para abruptamente desaparecer civiles.
En 1925, Sharpe Dunaway, un empleado de la Gaceta de Arkansas, alegó que los soldados en Elaine habían “cometido un asesinato tras otro con toda la deliberación tranquila en el mundo, ya sea demasiado despiadado para realizar la enormidad de sus crímenes, o demasiado borracho con la luz de la luna para dar un maldito continental “. … La información anecdótica sugiere que las tropas estadounidenses también participan en la tortura de afroamericanos para que confiesen y dan información.
Hoy, estamos viendo estas etapas iniciales exactas: unidades de combate, transporte militar y liderazgo que se dirige ilegalmente a las poblaciones civiles como amenazas militares.
Los titulares ahora describirán la construcción rápida y sistemática de infraestructura militarizada para la detención y deportación de masa, que se construye pieza por pieza a la vista. Reconocer esto como una señal de advertencia de algo mucho peor no es lo suficientemente alarmista por ninguna medida; Es un imperativo moral basado en el precedente histórico. Lo que es diferente hoy es cómo Palantir y su vigilancia doméstica rama peregrine operan algoritmos opacos inseguros de orgoritmo, como si Wall Street leyera “The Trial” de Kafka y pensó que era una guía para las nuevas empresas de unicornio.
El tiempo para sonar la alarma fue antes de las elecciones, antes de las órdenes ejecutivos, antes de la confirmación del Senado. Todavía existen algunos mecanismos críticos de supervisión, pero quién sabe si se quedará algo: los comités de supervisión del Congreso pueden exigir respuestas sobre despliegues de tropas y operaciones militares en suelo estadounidense. Los soldados pueden rechazar órdenes ilegales. Los fiscales generales estatales retienen la autoridad para impugnar la extralimitación federal. Las organizaciones de derechos civiles aún pueden presentar desafíos legales contra la detención militar. Los periodistas aún tienen protecciones de la Primera Enmienda para documentar y exponer estas operaciones.
La historia preguntará qué hicimos cuando vimos las señales claras. “America First” ha significado durante más de 100 años un terrorismo doméstico generalizado, un frente político para el KKK.
Y, sin embargo, aquí está nuevamente en el escenario federal como si todos lo olviden todo.
Qué supervisión exigimos, qué desafíos presentamos, qué historias documentamos, qué resistencia montamos. La respuesta no puede ser que miramos hacia otro lado mientras la infraestructura para la tragedia racista de los derechos humanos en masa se construyó a la vista, nuevamente.
-
Startups8 meses ago
Remove.bg: La Revolución en la Edición de Imágenes que Debes Conocer
-
Recursos9 meses ago
Cómo Empezar con Popai.pro: Tu Espacio Personal de IA – Guía Completa, Instalación, Versiones y Precios
-
Recursos9 meses ago
Suno.com: La Revolución en la Creación Musical con Inteligencia Artificial
-
Recursos8 meses ago
Perplexity aplicado al Marketing Digital y Estrategias SEO
-
Tutoriales9 meses ago
Cómo Comenzar a Utilizar ChatGPT: Una Guía Completa para Principiantes
-
Startups7 meses ago
Startups de IA en EE.UU. que han recaudado más de $100M en 2024
-
Estudiar IA8 meses ago
Curso de Inteligencia Artificial de UC Berkeley estratégico para negocios
-
Noticias7 meses ago
Dos periodistas octogenarios deman a ChatGPT por robar su trabajo