Mar 26, 2025 · 36m · a16z
The Top 100 GenAI Products, Ranked and Explained
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this episode of The a16z Podcast, partners Olivia Moore and Anish Acharya unpack insights from the fourth edition of the Gen AI 100 report, analyzing consumer adoption trends, high-growth entrants like DeepSeek, vibe coding platforms, and revenue strategies across mobile and web ecosystems.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. How this is scored →
The host as informed peer 3.9 Guest teaching 5.1 Guest disagreement 0.6 The host pushing back 0.2
speaking balance: gold is the host, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 7:17 Reframing AI work delegation assumptions
Anish contrarianly rejects the standard premise that humans will delegate to AI, proposing instead that AI will delegate work to humans.
Hardest push from the host ▶ 14:55 Probing deep research product directionHost pushes back on the assumed general use of deep research, pressing Anish on whether vertical developer applications are the more likely progression.
Biggest teaching moment ▶ 24:25 Detailed retention stats and geo VPN dynamicsOlivia educates the panel with exact Day 30 retention percentages for DeepSeek versus ChatGPT, adding geopolitical context around VPN usage in China.
The host holds their own ▶ 28:33 Host detailing personal Ideogram prompt workflowHost demonstrates strong domain expertise by recounting how she utilized Ideogram's text layer specificity during an internal AI video competition.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | The host as informed peer | Guest teaching | Guest disagreement | The host pushing back | Why |
|---|---|---|---|---|---|---|
| Tracking the Consumer AI Landscape | 3 | 5 | 0 | 0 | Host introduces the report and asks methodology questions. Guest Olivia Moore details how web and mobile datasets are pulled from SimilarWeb and Sensor Tower. | |
| Key Moments in the Consumer AI Timeline | 3 | 6 | 1 | 0 | Host asks for pivotal timeline moments. Olivia and Anish educate the host on pre-ChatGPT products like Midjourney and Character AI, as well as DeepSeek releasing its reasoning model for free. | |
| Challenging Assumptions in AI Product Development | 5 | 5 | 1 | 1 | Host offers a strong thematic framework about matching pivotal timeline moments with shattered assumptions. Anish counters standard assumptions by suggesting AIs might delegate work to humans rather than vice versa. | |
| Emergence of AI Video Models | 4 | 5 | 1 | 0 | Host quotes directly from the report regarding unexpected leaderboard shifters. Olivia and Anish explain the rise of Chinese AI video models and vibe coding platforms like Bolt and Lovable. | |
| Introducing The Brink List | 3 | 5 | 1 | 0 | Host asks about the new Brink List feature. Guests explain missing categories such as style transfer video and upcoming model primitives like Gemini Flash and Operator. | |
| Deep Research as an Application Primitive | 4 | 4 | 1 | 1 | Host double-clicks on Deep Research, pushing guests on whether it will remain horizontal or be tailored into niche developer tools. Olivia shares practical examples like tracing meme origins. | |
| Companion Multimodality and Creation vs. Consumption Dynamics | 4 | 5 | 0 | 0 | Host asks about ChatGPT's non-linear growth trajectory. Olivia provides data on ChatGPT's plateau and subsequent surge to 400M WAUs following reasoning and multimodal updates. | |
| Claude's Unique Position and DeepSeek Retention Dynamics | 3 | 6 | 1 | 0 | Anish positions Claude as a unique power-user model. Olivia educates on exact retention data (7% D30 for DeepSeek vs 9% D30 for ChatGPT) and geopolitical VPN factors. | |
| Global AI Video Ecosystem: Western vs. Chinese Models | 4 | 5 | 1 | 0 | Host asks if video models are becoming opinionated in features. Olivia explains promptable cinematography controls, while Anish outlines aggregator platforms like Crea. | |
| Ideogram and Specificity in Image Generation | 6 | 5 | 0 | 0 | Host demonstrates hands-on experience using Ideogram in an internal video competition to get precise text before animating. Olivia explains mobile revenue vs active user disparities. | |
| High-ARR Niche Applications vs. Mainstream Scale | 4 | 5 | 0 | 0 | Host prompts discussion on monetization trade-offs vs user growth. Olivia uses plant ID apps to illustrate high-ARPU niche strategies and warns against builder traps. |
Statements from this episode (41)
Acharya: App stores will become chaotic as AI lowers creation costs
“The app store is going to be chaos.”
Moore: Mainstream AI products often avoid marketing themselves as AI
“But actually there's a whole separate set of companies and products that might be reaching the true mainstream consumer that might not even be marketing themselves as AI products, but they're kind of powered by and made possible by AI.”
Moore: Consumer AI adoption lags research breakthroughs by 6 to 12 months
“Consumer activity typically lags by six to nine to 12 months, what's happening on the research side.”
Moore: Midjourney and Character.ai launched before ChatGPT in 2022
“Actually, Midjourney and Character AI both came out before ChatGBT, which I think a lot of people don't know, but there was maybe these early niche communities of early adopters that were using both of those products in the summer and the fall of twenty-twenty…”
Moore: Around 150 million people used Snapchat's MyAI chatbot
“And then post ChatGPT, things that just brought AI to consumer consciousness, so even remember Snapchat's MyAI with that little bot that appeared at the very top of your feed, and like a hundred fifty million people used it, and for a lot of kind of younger co…”
Acharya: DeepSeek offered reasoning models for free vs OpenAI's paid tier
“So, one I think important nuance is the fact that they released their reasoning model for free at scale, so previously you had to use O-one Pro, and you had to pay ChatGPT's premium subscription to get access to it.”
Acharya: Real-time chain-of-thought display became an industry standard after DeepSeek
“It showed you its thought process in real time. It was just super captivating, and now something that's become a step that every model takes.”
Moore: AI video will eventually create full-length movies
“AI video right now can generate great three or five or six second clips, but hopefully years from now we have AI video that can generate minutes long or, you know, even hours long movies.”
Acharya: AI may end up delegating work to humans
“Like, what if the AIs are the ones delegating the work to us? Perhaps AI is really good at organizing work and we're really good and also get a lot of joy out of doing it. So I think so many of these assumptions and that's why their assumptions, they seem intu…”
Moore: a16z GenAI 100 report added three new video models including Sora
“We had three new video models on the list this time Hyluo and Kling, which are both Chinese models, and then Sora, which was OpenAI's model that was announced, I guess, more than a year ago at this point, and finally was released.”
Moore: Google's Veo 2 video model likely releasing in 3-6 months
“I think we'll see even more of a shake-up here, because VO-II is the new Google model that is even next level beyond that, from what we've seen in testing, and that is probably finally gonna hopefully come out in the next, you know, three or six months.”
Acharya: Vibe coding enables disposable software with hyper-short shelf lives
“These products make it possible to create a product or an experience that may have an extremely short shelf life, like, 20 minutes or a week or, you know, any other time period.”
Moore: Runway, Otter, and UMax dropped from Top 50 GenAI list
“There were a couple companies on the list, like Runway, Otter, UMax, across web and mobile, that have been on the core top 50 ranks in the past. But maybe they got just edged out by, like, Deep Seek launching this time, and so they lost their spot for this ran…”
Moore: KREA and Lovable may soon reach Top 50 GenAI list
“And then the other trend that it caught was a rise in more recent products, like Crea made the list, and Lovable made the list, that are very much on the kind of consistent upswing, and if it continues, we might see them on the main ranks, and they haven't mad…”
Acharya: Style transfer has lower inference cost than text-to-video
“Style transfer is just a much more tractable, tractable problem and has a lot lower cost of inference versus raw text to video.”
Moore: Agentic and voice AI apps will explode in upcoming rankings
“I think there's always a lag because the models have to be released to developers, and they have to be tuned by the developers, and so it takes a while, but I would expect to see Maybe an explosion of fun and unique and interesting products built on models lik…”
Acharya: AI deep research is a primitive, not a standalone product
“Deep research which if you played with it is completely magical, but it's a primitive, right? It's not a product. It's something to build other things with.”
Acharya: Constrained apps built on OpenAI Deep Research are underexplored
“Not more likely, but I think it's underexplored. If you come to deep research today, you kind of have the blank page problem, and I'd love to see developers, you know, create some constraints that leads to unexpected outcomes.”
Moore: OpenAI Deep Research is 100x better than Know Your Meme
“But if you try other things, like one day we were trying to trace the origin of a meme, and deep research is, like, a hundred-x better version of that Know Your Meme website that kind of, like, goes through the history.”
Moore: Around 95% of YC startups build using AI development tools
“95% of YC companies or something are now Kind of building using those tools.”
Moore: Three AI companion products ranked in the GenAI 100 top 10
“I think we had three companion products in the top 10. Two of them were NSFW oriented.”
Moore: 16 companies appeared on all four GenAI 100 web rankings
“Across the four lists, there's now 16 companies on the web ranks who have made it every single time and have kept the streak going”
Acharya: AI companion app demand will surge with multimodality
“There's so much latent demand and it'll really increase Once you have multi-modality.”
Moore: Lovable creator domain traffic far outweighs visits to deployed apps
“And, like, lovable.dev has more usage or more visits significantly than, you know, traffic to lovable.app, which gets back to what I was saying before.”
Moore: First wave of viral apps from vibe-coding platforms yet to come
“Like, we have not even seen the first wave of viral products built on top of lovable and bold. And so when that happens, I think that the awareness of these types of platforms is gonna go significantly up, too.”
Moore: ChatGPT web visits remained flat from February 2023 to February 2024
“Between February, 20, 23, basically for a whole year through February, 20, 24, it was essentially flat in monthly visits to the website.”
Moore: ChatGPT doubled weekly active users to 400M in six months
“In the past six months, they grew from two hundred million to four hundred million weekly active users.”
Acharya: Anthropic's Claude surpasses ChatGPT in coding and creative writing
“Claude sits in this very interesting place where it seems like it's more beloved by a smaller number of people. It's better at, Creative writing. You know, it seems to have more of a personality, which is interesting, because at least I think it's designed to …”
Moore: DeepSeek reached #2 AI web product within a week of launch
“We only had 10 days of data for Deep Seek for January, because it launched at the end of the month, and it shot up from literally nothing to number two on the list, you know, 10% of ChatGPT scale on web within a week, a little bit more than a week.”
Moore: DeepSeek mobile app achieves 7% 30-day retention versus ChatGPT's 9%
“On a retention basis, like, how many users are coming back to the app, say, Exactly 30 days, exactly seven days, exactly 60 days. It's just slightly below ChatGPT. So we're looking at seven percent day 30 for DeepSeq and nine percent day 30 for ChatGPT.”
Moore: Two top AI video models on GenAI 100 list are Chinese
“I mean, two of the video models were Chinese video models, which is super interesting.”
Moore: Chinese AI video models rely on less copyright-sensitive training data
“The models are less copyright sensitive in their training data.”
Moore: Chinese AI video models outperformed expectations relative to capital raised
“Sora in some ways was a little bit disappointing for some people in, and whereas like the Chinese video models were maybe better than a lot of people expected, given the relative lack of capital that they've raised.”
Acharya: KREA is the single best platform to access all AI models
“You know, CREA is the single best place to access all the models and all the tools, and the nice thing that they do is stitch all of these things together to make them greater than the sum of their parts.”
Acharya: Ideogram remains one of the most unique AI image models
“I mean, the comments specifically, I still think Ideogram is one of the most unique models for sort of what it does, what it's great at, which is text generation, sort of aesthetic that it has.”
Moore: Voice-first AI products succeed on mobile over web
“A lot of the voice-first consumer products that we are seeing working actually are on mobile versus web because it's easier and more natural to talk into your phone for language learning or for companionship or other use cases than it is to maybe talk into you…”
Moore: Homework helper AI apps are blowing up on mobile
“And same with homework helper apps. Like, those are really blowing up on mobile as compared to web.”
Moore: Top AI mobile apps by usage and revenue only overlap 40%
“So for the first time, we actually ranked the top 50 by What Sensor Tower can measure as mobile revenue, which is typically in-app purchases and subscriptions, so probably not ads. And we ranked those separately from what has the most monthly active users. And…”
Moore: Smaller AI mobile apps generate significantly higher revenue per user
“We actually found, we plotted like revenue per user versus number of users. And we found the apps that had smaller user bases out of this sample set We're much more likely to be making significantly more money on a per user basis.”
Moore: Plant ID apps ranked high in mobile AI revenue, not usage
“One example of a category that did appear on the mobile revenue list, but was not on mobile usage, Was several plant identification apps.”
Moore: Top PhD AI researchers often struggle building consumer products
“Often we meet with these amazing, like, PhD researchers, best in class in the whole world in terms of their technical understanding of a model or a capability and they can struggle building in consumers sometimes because often the more complicated thing is not…”