See every source

Benchmark source register

Inspect every benchmark publisher, methodology link, update date and coverage boundary used by TapToMarket.

Source-first. No hidden ranking weight.TapToMarket did not run these tools or benchmarks. We organize external evidence and state where it applies. Affiliate commission, sponsorship and EPC carry 0% weight.

See every source

8 sources

Methodology
AI video generators

Artificial Analysis Video Arena

A live arena leaderboard for video-generation models. It can inform model choice but does not score complete avatar-video products.

Source
Artificial Analysis
Checked
Live source
Evidence coverage
Official source-hosted embed
AI video generators

VBench Leaderboard

A multi-dimensional video benchmark covering visual quality, motion and consistency.

Source
VBench team
Checked
2026-04-01
Evidence coverage
MIT space · source assets may differ
AI voice generators

Artificial Analysis Speech Arena

Listener preference data for speech providers, presented in the publisher’s live view.

Source
Artificial Analysis
Checked
Live source
Evidence coverage
Official source-hosted embed
AI voice generators

TTS Arena V2

A second listener-based view of naturalness and preference across TTS models.

Source
TTS-AGI & contributors
Checked
2026-09-05
Evidence coverage
Apache-2.0 space
AI coding tools

BigCodeBench

Measures model performance on multi-library programming tasks. It is useful context, but it does not rank Cursor, Windsurf or Bolt as products.

Source
BigCode project
Checked
2025-02-04
Evidence coverage
Apache-2.0 space
AI image generators

Artificial Analysis Image Arena

A live preference leaderboard for image-model outputs. Product workflow and editing control still come from separate review evidence.

Source
Artificial Analysis
Checked
Live source
Evidence coverage
Official source-hosted embed
AI meeting assistants

Open ASR Leaderboard

Plots word error rate against speed across ASR models. It does not measure meeting summaries, action items or integrations.

Source
Hugging Face & contributors
Checked
2026-09-11
Evidence coverage
Apache-2.0 app · component datasets vary
AI coding tools

DeepSWE

Measures frontier coding agents on 113 original, long-horizon engineering tasks across 91 repositories and five languages.

Source
Datacurve
Checked
2026-09-03
Evidence coverage
Publisher-hosted public leaderboard