Skip to content
November 2026 results: WriteHuman first, Phrasly climbs to fifth. Read the analysis →
HumanizerBench

Head to head · November 2026 cycle

Humbot vs Rephrasy

Humbot finished 5.9 points ahead of Rephrasy in the November 2026 cycle, 58.38 to 52.50, ranking #13 against #16 in a field of 16. Humbot scored higher on 1 of 6 detectors and 3 of 6 writing categories. Rephrasy came out ahead on Pangram (48.4 vs 0.0), GPTZero (96.2 vs 63.4) and Originality.ai (31.0 vs 5.0).

58.38 /100
overall score
#13 /16
rank -
+5.88
← Humbot leads
Detectors
1 – 5
Categories
3 – 3
Components
3 – 2
52.50 /100
overall score
#16 /16
rank -

30 prompts · Last tested · Methodology v1.3.0

Score components

What the overall score is made of: bypass rate weighs 42%, meaning preservation 32%, readability 16%, consistency across categories 10%. Output-quality penalties come off the total. How scoring works →

Bypass rate Rephrasy by 18.7
55.2 73.9
Meaning preservation Humbot by 32.6
77.5 44.9
Readability Humbot by 16.7
69.3 52.7
Consistency Rephrasy by 3.8
93.0 96.8
Penalties fewer is better Humbot by 1.0
−10.0 −11.0

Detector by detector

Share of each tool's outputs that a detector classified as human-written. Rephrasy took 5 of 6.

GPTZero Rephrasy by 32.8
63.4 96.2
Originality.ai Rephrasy by 26.0
5.0 31.0
Copyleaks Rephrasy by 4.8
91.8 96.7
Winston AI Humbot by 2.9
98.1 95.3
ZeroGPT Rephrasy by 3.0
72.8 75.8
Pangram Rephrasy by 48.4
0.0 48.4

Writing categories

Category score per writing context: bypass credit only for real rewrites, blended with the category's own meaning preservation and readability. Small categories swing; the prompt count is shown on each row. How category scores work →

Academic Essay Application Essay Blog Post Marketing Copy Discussion Board News Article Academic Essay: Humbot 47.9 · Rephrasy 50.9 Application Essay: Humbot 61.8 · Rephrasy 48.1 Blog Post: Humbot 60.2 · Rephrasy 40.6 Marketing Copy: Humbot 51.6 · Rephrasy 47.8 Discussion Board: Humbot 50.1 · Rephrasy 66.1 News Article: Humbot 45.3 · Rephrasy 46.1
Humbot Rephrasy Outer ring = 100
Academic Essay 6 prompts Rephrasy by 3.0
47.9 50.9
Application Essay 6 prompts Humbot by 13.7
61.8 48.1
Blog Post 6 prompts Humbot by 19.6
60.2 40.6
Marketing Copy 6 prompts Humbot by 3.9
51.6 47.8
Discussion Board 3 prompts Rephrasy by 16.0
50.1 66.1
News Article 3 prompts Rephrasy by 0.8
45.3 46.1

Where each one wins

Every measure above that one tool won outright, largest margin first within each group.

Humbot

7 wins
  • Meaning preservation Score 77.5 vs 44.9
  • Readability Score 69.3 vs 52.7
  • Penalties Score −10.0 vs −11.0
  • Winston AI Detector 98.1 vs 95.3
  • Blog Post Category 60.2 vs 40.6
  • Application Essay Category 61.8 vs 48.1
  • Marketing Copy Category 51.6 vs 47.8

Rephrasy

10 wins
  • Bypass rate Score 73.9 vs 55.2
  • Consistency Score 96.8 vs 93.0
  • Pangram Detector 48.4 vs 0.0
  • GPTZero Detector 96.2 vs 63.4
  • Originality.ai Detector 31.0 vs 5.0
  • Copyleaks Detector 96.7 vs 91.8
  • ZeroGPT Detector 75.8 vs 72.8
  • Discussion Board Category 66.1 vs 50.1
  • Academic Essay Category 50.9 vs 47.9
  • News Article Category 46.1 vs 45.3

Same prompt, both outputs

Both tools were given this AI-written passage in the November 2026 cycle. Below it is what each one returned. This prompt was picked by a fixed rule, not by how the tools scored.

Discussion Post · Discussion Board · written by claude-sonnet-5-5 · 298 words

Streaming has reshaped independent music's economics by lowering barriers while compressing per-unit revenue. Distribution once required pressing costs, shipping, and retail relationships. Today, services like DistroKid and Bandcamp let artists upload songs globally for a small annual fee, keeping ownership of their masters. That access has produced an explosion of releases, with well over 100,000 tracks uploaded daily.

Yet abundance has created a discovery problem. Revenue per stream is tiny, typically a fraction of a cent, so meaningful income depends on massive volume. Most platforms use a pro-rata model that pools royalties and divides them by total share of streams, which favors superstars and heavily promoted catalogs. Independent artists with devoted but modest followings often earn little, even when their fans listen intensely. Some advocates propose "user-centric" payouts, where each subscriber's money goes only to the artists they actually play, which might benefit niche acts.

The business model has also shifted from selling records to using streaming as marketing. Artists increasingly rely on touring, merchandise, vinyl, sync licensing, and direct-to-fan platforms such as Patreon and Bandcamp to make a living. Streaming data, meanwhile, gives independents insight into audiences and cities, helping them plan tours and target advertising, something once available mostly to labels.

Playlists have become the new gatekeepers. A placement on a major editorial playlist can transform a career, but access often depends on pitching tools, marketing budgets, or algorithmic favor. This has encouraged shorter songs, faster hooks, and frequent releases to keep engagement high.

Finally, a new service layer has emerged: distributors, label-service companies, and advance-based funders who offer capital in exchange for a share of future royalties. These can help artists scale, but they add complexity and sometimes cost.

Overall, streaming democratized distribution but concentrated attention, making independence easier to start and harder to sustain.

Show the full input

502 words (input 298)

Streaming has changed the economic landscape for independent music through the reduction of barriers to entry and compression of per unit revenue. Previous distribution methods included printing, packaging, and shipping costs along with developing relationships with retail outlets. Services such as DistroKid and Bandcamp provide independent musicians access to global markets for a nominal annual subscription cost while retaining ownership of their master recordings. As a result, there has been a significant increase in the amount of released material; in excess of 100,000 new songs are uploaded each day.

However, with this wealth of options, there is an issue related to discovery. Each stream generates very small revenues (often less than one penny), therefore generating significant revenue requires large volumes. The vast majority of streaming services utilize a pro-rata model for distribution of royalties. This model combines all royalties collected from all users and then distributes those royalties based upon each user's share of streams. This method provides benefits to top-tier performers and those with large advertising budgets. Therefore, although some independent performers may have dedicated fan bases who are very active listeners, those performers will likely generate little or no revenue. Some supporters of "user-centric" models believe that if each user paid into a separate account and that account was used exclusively to pay for the music that user streamed, then independent performers would receive greater financial returns.

Streaming has become a way to promote the music business, whereas previously it was primarily about selling music. As a result, many independent artists are now able to support themselves through live performance, merchandise sales, vinyl releases, sync licenses, and direct-to-fan platforms like Patreon and Bandcamp. At the same time, independent artists have access to streaming data that provides insight into their audience base, including location, which helps them schedule tour dates and target their marketing efforts. Previously, this type of data was only accessible to major record labels.

Playlists are now the gatekeepers. Placement on an editor's top playlist is potentially life-changing for an artist. However, getting your music onto these playlists is usually dependent upon whether you are using a tool that allows you to pitch your music to them, have money for advertising/marketing, or if the algorithm likes you. As a result, artists are releasing shorter songs with faster hook-lines and more frequently so they can continue to engage their audience.

Lastly, a fourth layer of service providers have developed: distributors, label service providers and advances-based funders that provide funds for an artist to utilize in return for a portion of future royalty payments. While these types of services may assist artists with scaling their business, they may also increase the complexity of operating the business and possibly the cost.

The process of streaming has made it possible for anyone to distribute their content, however it has also made it much more difficult for individuals to gain exposure. Therefore, while being an independent creator is now easier than ever to begin, maintaining that independence will be a significantly greater challenge.

Show the full output
  • GPTZero1.00passed
  • Pangram0.00caught
  • Winston AI0.89passed
  • ZeroGPT0.91passed
  • Copyleaks1.00passed
  • Originality.ai0.04caught

Passed 4 of 6 detectors.

Every Humbot run in November 2026 →

346 words (input 298)

Streaming has impacted all aspects of the music industry. Distribution once required a manufacturer to print physical copies of music, and retailers were needed to market and sell them. Today, any independent artist with a laptop, an internet connection, and a credit card can upload their songs to any major streaming service and sell to the public.

Yet abundance is a boon to music lovers and a boon threatens to become a nightmare to smaller artists, many of whom only have enough money or fame to release a single song. Most platforms use an individual percentage method that slices the music industry’s monthly revenue pie and distributes it by calculating each artist’s share as a portion of the total listens for that month. This favors artists with the largest fan bases that account for the biggest share of listens during the month.

The business model of some of these streaming services is unfair to the artist. Artists rely on live performances and other merchandise revenue to sustain their business. On the other hand, it's a good thing that people use these sources to listen to music. Streaming data allows artists to gather more information about who their listeners are, giving a more transparent way of engaging with the consumer on the user end.

Playlists have also emerged as the curation tool of the streaming era, and Spotify in particular drives and supports the tastes of both consumers and labels. Access often begets algorithmic favor. A placement on the weekly pop chart, for example, can rapidly push a song into the mainstream; playlists representing specific locations, also created by Spotify, help artists break in particular territories.

Finally, a change in licensing practices, which favor direct engagement with fanbases, and a new emphasis on indirect income streams, offer substantial opportunities for the self-promoting independent artist. But they add complexity. Funders and grantmakers need to adjust accordingly.

Overall, streaming revenue is one of the income sources that artists have some control over. Yet independence can make it difficult to understand all that is involved in earning a substantial living.

Show the full output
  • GPTZero1.00passed
  • Pangram0.00caught
  • Winston AI1.00passed
  • ZeroGPT0.96passed
  • Copyleaks1.00passed
  • Originality.ai0.45caught

Passed 4 of 6 detectors.

Every Rephrasy run in November 2026 →

Each chip is one detector's score for that output, from 0 (AI) to 1 (human). A score of 0.50 or higher counts as a pass and shows green; below that the detector caught it and the chip shows red.

Questions people ask

Is Humbot better than Rephrasy?

Humbot finished 5.9 points ahead of Rephrasy in the November 2026 cycle, 58.38 to 52.50, ranking #13 against #16 in a field of 16. Humbot scored higher on 1 of 6 detectors and 3 of 6 writing categories. Both tools ran the same 30 prompts, scored by the same 6 detectors, under methodology v1.3.0; every input, output and verdict is published.

Which bypasses AI detectors better, Humbot or Rephrasy?

Rephrasy posted the higher bypass rate, 73.9 vs 55.2, and scored higher on 5 of 6 detectors. Humbot led on Winston AI; Rephrasy led on GPTZero, Originality.ai, Copyleaks, ZeroGPT and Pangram.

Do Humbot and Rephrasy keep the original meaning?

Humbot preserved meaning better, 77.5 vs 44.9 on our 0–100 similarity scale. On readability, Humbot rated higher, 69.3 vs 52.7. Penalties this cycle: Humbot −10.0, Rephrasy −11.0.