A composite score ranking AI companies by real developer traction - not hype, not funding, not press mentions.
The Developer Momentum Index is AI-Buzz's proprietary score that measures how much traction an AI company has with developers. Unlike funding-based rankings, the DMI reflects actual adoption and community engagement.
Every company in the AI-Buzz database receives a DMI score from 0 to 100, updated daily. The score aggregates signals that traditional databases miss: who's actually installing packages, who developers are discussing, and which projects are accelerating.
Scores updated daily. View full rankings →
npm and PyPI package downloads (30-day rolling totals), GitHub stars, forks, and contributor counts. These reflect who developers are actually using in production code.
Hacker News mentions (30-day count), category discussion share, and sentiment analysis. HN discussion is a leading indicator - what developers talk about today often becomes mainstream adoption 6-12 months later.
Growth trends across all metrics - month-over-month download growth, star velocity, commit activity acceleration. A company growing at 20% monthly from a small base can outrank a stagnant giant.
Google Search Console impressions and clicks for company-related queries. This captures mainstream awareness beyond developer communities.
Traffic to each company's AI-Buzz profile. Higher interest from researchers and analysts correlates with broader industry attention.
Each signal is normalized to 0-1 before weighting. Weights sum to 100%.
| Signal | Weight |
|---|---|
| Supply-side signals - 40% | |
| HN Mentions | 15% |
| GitHub Stars | 10% |
| Download Momentum | 10% |
| Funding Recency | 5% |
| Demand-side signals - 40% | |
| Page Views | 15% |
| Search Appearances | 10% |
| GSC Impressions | 15% |
| Data quality - 20% | |
| Data Coverage | 10% |
| Package Downloads | 10% |
Strong developer adoption, active community discussion, and growing traction. These companies are the ones developers are actively installing and talking about.
Moderate developer engagement. May have strong adoption in a niche or early-stage growth across multiple signals.
Lower developer traction relative to other tracked companies. May be pre-launch, enterprise-focused, or in a quieter growth phase.
All signals are collected daily at 6 AM UTC via automated GitHub Actions workflows. See our full methodology for details on each data source.
Single metrics can be misleading. A company with millions of downloads but zero community discussion is different from one with moderate downloads but explosive growth. The DMI combines multiple independent signals to give a more complete picture of developer traction.
Yes. DMI data is available via the AI-Buzz API and as CSV/JSON exports. Please cite AI-Buzz as the source and link to this page for methodology context.
A company needs at least one quantitative signal (downloads, stars, HN mentions, etc.) to receive a DMI score. Companies without any public developer signals will show "No Data" until signals are detected.
Hacker News discussion is a stronger leading indicator of future adoption than cumulative star counts. HN mentions reflect active developer attention in a 30-day window — what developers debate today often becomes mainstream adoption months later. GitHub stars accumulate over a project's lifetime and can reflect historical popularity rather than current momentum.
The full weight breakdown: supply-side signals (HN mentions 15%, GitHub stars 10%, download momentum 10%, funding recency 5%) account for 40%. Demand-side signals (page views 15%, search appearances 10%, GSC impressions 15%) account for 40%. Data quality (coverage 10%, package downloads 10%) accounts for 20%. Each signal is normalized against the maximum value across all tracked companies using max-based normalization (value / max).
Funding recency uses a decay function: funding within the last 30 days contributes the full 5% weight (boost factor 1.0), 31–90 days gets 0.7, 91–180 days gets 0.4, 181–365 days gets 0.2, and anything older than a year gets 0.05. This ensures recently funded companies get a modest boost without dominating the score.
Weights are organized into three categories: Supply-Side (40%), Demand-Side (40%), and Data Quality (20%). Supply-side signals measure what developers produce — code contributions, package publications, and community discussion. Demand-side signals measure what developers consume — page views, search queries, and search impressions. Data quality signals reward companies with more complete data coverage. Weights were set based on signal reliability, independence from each other, and predictive value for sustained developer adoption.
Developer adoption signals like page views, search appearances, and GSC impressions reflect real-world interest from developers actively evaluating tools. A developer searching for a company or visiting its profile indicates active consideration, which is a distinct signal from supply-side production metrics like downloads or stars. Giving demand-side equal weight to supply-side ensures that companies generating genuine developer interest are recognized even if they have smaller open-source footprints.
The DMI uses 9 weighted components across 3 categories:
DMI weights are reviewed quarterly. Recalibration considers whether any signal has become gameable, whether new data sources should be incorporated, and whether the current weights still correlate with sustained developer adoption. Any weight changes are documented on this page and in the data changelog.
Import DMI data directly into Google Sheets with IMPORTDATA():
Fetch JSON data for Airtable scripting or any API client:
Filter by domain: ?primary_domain=foundation-models. Filter by score: ?min_signal_score=0.5. Rate limited to 10 requests/hour.
See which AI companies have the strongest developer momentum right now.
Which AI companies are developers actually adopting? We track npm and PyPI downloads for hundreds of companies. Get the biggest shifts weekly - before they show up in the news.