State of Product Launches, October 2026: 39,000 Launches by the Numbers
October 31, 2026 · Nicheloom · 7 min read · Data reports
In October 2026, about 39,000 new products launched across Product Hunt, GitHub, Hacker News and Y Combinator, roughly 1,260 a day. 27% were AI-native, 68% were commercial products, and the biggest single niche was AI video generation tools with 371 launches.
These numbers come from Nicheloom's own daily tracking, for the 30 complete days from October 1 to October 30, 2026. Spam, lures and bait listings are removed first (more on that at the end), and every table says how many launches it rests on.
October at a glance
| Question | Answer | Based on |
|---|---|---|
| Launches in the month | 39,291 | Oct 1 to 30, after cleaning |
| Launches per day (median) | 1,264 | 30 days (range 932 to 1,735) |
| Busiest weekday | Tuesday, 1,694 | median per weekday |
| Quietest weekday | Sunday, 1,025 | median per weekday |
| AI-native launches | 27% | 38,297 launches assessed |
| Commercial products | 68% | 38,131 assessed |
| Could be built with AI coding tools | 61% yes, 24% partly | 39,289 assessed |
| Niches with at least one launch | 814 | median niche had 35 launches |
Where launches came from
| Source | Launches | Share |
|---|---|---|
| Product Hunt | 28,455 | 72.4% |
| GitHub | 10,138 | 25.8% |
| Hacker News | 581 | 1.5% |
| Y Combinator | 117 | 0.3% |
Product Hunt dominates the count, so the weekday pattern above mostly reflects Product Hunt, which is busiest midweek. Hacker News is small because we only collect posts that reached a minimum number of points, and GitHub is the newest repos we found, not all of GitHub.
How much of it is AI?
About 27% of launches were built around AI, a further 10% used AI as a feature, and 64% did not use AI as a core part of the product. The share changes a lot by source:
| Source | AI-native share |
|---|---|
| Y Combinator | 82% |
| Hacker News | 34% |
| Product Hunt | 26% |
| GitHub | 26% |
Y Combinator is the outlier: 82% of its listed companies are AI-native, against about a quarter of everything else.
It also changes by industry. Among commercial products with a stated industry:
| Industry | AI-native share | Launches |
|---|---|---|
| Sales | 43% | 525 |
| HR | 39% | 726 |
| Marketing | 36% | 2,015 |
| Media | 33% | 3,869 |
| Education | 27% | 1,267 |
| Healthcare | 22% | 1,028 |
| Fintech | 15% | 1,673 |
| Travel | 11% | 652 |
AI is most common where the work is writing, outreach and reports, and least common in travel and fintech.
Who launches are built for, and what they do
| Built for | Share |
|---|---|
| Consumers | 36% |
| Prosumers (power users and professionals) | 25% |
| Developers | 19% |
| Businesses | 19% |
| What it does | Share |
|---|---|
| Workflow and automation | 20% |
| Developer tools | 19% |
| Content generation | 14% |
| AI agents | 9% |
| Analytics | 8% |
| Software for one industry | 7% |
Workflow tools and developer tools are the two biggest categories. Two thirds of launches (68%) are commercial products, and one third (32%) are hobby or open source projects, which is mostly GitHub.
The biggest niches
Our model groups every launch into a niche. These were the twelve biggest in October:
| Niche | Launches | First half | Second half |
|---|---|---|---|
| AI video generation tools | 371 | 186 | 185 |
| Video game source ports | 314 | 163 | 151 |
| Browser-based utility tools | 305 | 157 | 148 |
| PC gaming optimization utilities | 260 | 118 | 142 |
| Developer utility tools | 244 | 156 | 88 |
| Specialized calculation tools | 238 | 105 | 133 |
| Language learning apps | 226 | 98 | 128 |
| AI coding agent session managers | 224 | 100 | 124 |
| Model context protocol tools | 200 | 121 | 79 |
| Video game mods and managers | 199 | 102 | 97 |
| Fitness and health trackers | 199 | 93 | 106 |
| Generative engine optimization software | 197 | 103 | 94 |
Most of the biggest niches were steady across the month. Two slowed noticeably: developer utility tools (156 to 88) and model context protocol tools (121 to 79). The first half is October 1 to 15 and the second half is October 16 to 30.
No single niche dominates. The ten biggest together hold only 6.6% of all launches. Of 814 niches with a launch, 272 had 50 or more and 70 had 100 or more. Browse them on the niche list.
What the most popular new projects were
On GitHub, the most-starred new projects with a description:
| Project | Stars | What it does |
|---|---|---|
| cleanyourtv | 17,590 | Removes ads and bloat from Android and Google TV |
| DeepInfer | 10,428 | Inference framework for DeepSeek models |
| anatomy | 5,458 | An agent skill for interactive isometric illustrations |
| wordcraft | 4,832 | Microsoft Word reimplemented in Rust |
| aerial-tv | 4,312 | An ad-free, tvOS-style home screen for Android TVs |
| android-ad-cleaner | 4,223 | Removes ads and marketing pushes from Chinese-brand Android phones, no root |
| free-my-tv | 3,806 | An ad-free home screen for Google TV, LG webOS and Samsung Tizen |
Four of the top nine were about removing ads from TVs and phones. Only DeepInfer and anatomy are AI-native in the top group. Several of these projects have Chinese-language descriptions, so some of that pattern may be specific to Chinese phones and TVs. Star counts can be inflated, so treat them as attention, not quality.
On Hacker News, the most-upvoted new launches were:
| Launch | Points | What it does |
|---|---|---|
| Bigwords.page | 720 | Turns any screen into a sign |
| TitleArt | 408 | Makes text like it is 1997 |
| Llmspeed | 329 | Shows how fast a local LLM runs before you buy hardware |
| Unfenced | 317 | Land for sale that borders public land |
| WallHop | 286 | A replacement for a shut-down paywall viewer |
| Sphinx | 252 | A hook that quizzes you on your AI's code before shipping |
Almost none of these are conventional business software. Hacker News rewarded small, specific and playful projects.
What to do with these numbers
- Check your idea's niche size. The typical niche had 35 launches this month. Use Check your idea to see where yours falls.
- Do not assume AI means less competition. 27% of launches are AI-native, but industries like travel (11%) and fintech (15%) have far fewer, so there is less AI competition there.
- Look at steadiness. A niche with steady weekly launches is a more reliable signal than one that jumped in a single week. For how to judge this, read how to check if a startup niche is crowded.
- Remember the audience. Product Hunt skews toward consumer and prosumer products, Hacker News toward playful and technical projects. Pick the platform that fits your product.
How we counted
- Source: Nicheloom tracks launches from Product Hunt, GitHub, Hacker News and Y Combinator, 30 complete days from October 1 to October 30, 2026. Data was pulled on October 31, 2026.
- Cleaning: we removed 1,071 listings (2.7% of 40,362) flagged as bait or spam. These include repos named after paid software that promise cracked versions, Windows and Office activation tools, abuse tools, the same repo description copied by 8 or more repos, and gift card giveaway listings on Product Hunt. A few may slip through and a few real projects may be removed.
- Coverage: Product Hunt and Y Combinator are collected as fully as their sources allow. GitHub is the newest repos we found, sorted by stars, so it favors repos that gained attention. Hacker News only includes posts above a minimum number of points.
- Labels: AI-native, commercial, audience, function, industry and "could be built with AI coding tools" are assigned automatically by an AI model, so treat them as estimates. The counts behind each share are shown in the tables.
- Weekdays: daily counts per weekday are the median of the days in October.
- No September comparison: our September collection was still being expanded, so this series starts with October.
Data as of October 31, 2026, from Nicheloom's own tracking. This is the first post in a monthly series.
Frequently asked questions
How many products launch in a month? About 39,000 in October 2026 across Product Hunt, GitHub, Hacker News and Y Combinator, or around 1,260 a day, after removing spam and bait listings.
What share of new launches are AI? 27% were AI-native in October 2026, and another 10% used AI as a feature. 64% did not use AI as a core part of the product. Y Combinator's listed companies were 82% AI-native.
What is the busiest day for launches? Tuesday had the most launches in October, with a median of 1,694. Sunday was the quietest at 1,025. Product Hunt makes up about 72% of the count, so the pattern mostly follows Product Hunt.
What were the biggest startup niches in October 2026? AI video generation tools (371 launches), video game source ports (314), browser-based utility tools (305), PC gaming optimization utilities (260) and developer utility tools (244).
Which industries have the least AI? Among commercial products, travel (11%), fintech (15%), manufacturing (15%), construction (16%) and commerce (16%) had the lowest AI-native shares in October.