FreqBlog vs ReccoBeats
FreqBlog vs ReccoBeats: an honest comparison
Both APIs exist to fill the same gap: Spotify deprecated its public audio-features endpoint, and teams still need tempo, key, energy and mood for real tracks. The honest question a buyer asks is: “Which one fits how I actually identify tracks, and can I trust who is behind it?” This page sticks to publicly verifiable facts from each product’s own site and docs, and points you to ReccoBeats where it is genuinely the better fit.
At a glance
| Dimension | FreqBlog | ReccoBeats |
|---|---|---|
| How you resolve a track | By name + artist, ISRC, MusicBrainz ID, or Spotify ID — no Spotify ID required. | Docs state /track?ids= “supports both ReccoBeats IDs and Spotify IDs” (a Spotify Base-62 string or a ReccoBeats UUID); no track-name or ISRC lookup is documented. |
| Features without owning the audio | Look up by name/ISRC/ID with no audio upload. | Catalogue features are fetched by track ID; to analyse a track outside the catalogue you upload the audio file yourself to POST /v1/analysis/audio-features. |
| On-miss backfill | Auto-ingests an unmatched track on first request, then returns computed features. | No on-miss ingestion is documented; the audio-features endpoint lists 404 for tracks that are not found. |
| Fields returned | 42 fields including BPM, musical key (name + Camelot + Open Key), energy, mood, genre, ISRC and MBID. | Documented Spotify-parity set: acousticness, danceability, energy, instrumentalness, liveness, loudness, speechiness, tempo, valence. |
| Harmonic / DJ set-building | Camelot-aware set tools: /transition scoring, /next-track and /setlist ordering. |
Offers a track-recommendation endpoint (/v1/track/recommendation) based on song attributes; no Camelot harmonic set-builder is documented. |
| Hosted MCP server for LLMs/agents | Hosted MCP endpoint at mcp.freqblog.com/mcp. |
No hosted MCP server is documented on the site. |
| Operator disclosure | Operated by Steven Birring, Nuneaton UK, ICO-registered — a sole trader trading as Stackbase/FreqBlog. | Footer credits “Copyright © 2025 LatteBits”; no individual operator, registered business identity, or jurisdiction is named. |
Every ReccoBeats cell above reflects its own public site and documentation at the time of writing; product details can change, so verify against the live docs.
When ReccoBeats is the better fit
ReccoBeats is a good choice in a few concrete situations:
- You want a free API. ReccoBeats states it offers “completely free access” to its API. If an unlimited-at-zero-cost option matters more than named support or a monthly bill, that is a real advantage over FreqBlog’s paid tiers.
- You already live in Spotify IDs. If your data is keyed on Spotify track IDs and you just want a drop-in, Spotify-parity feature vector, ReccoBeats accepts those IDs directly.
- You already hold the audio. If you have the audio files on hand and want to extract features from your own uploads, ReccoBeats’
/v1/analysis/audio-featuresextraction endpoint is built for exactly that.
Why teams pick FreqBlog
FreqBlog is built for teams that identify music the way people actually talk about it — by name and artist — and that want to know who stands behind the data.
- Name-first lookup. Resolve a track by name + artist, ISRC, MusicBrainz ID or Spotify ID. You never have to map your library to Spotify IDs first.
- On-miss backfill. When a track isn’t in the ~576k-track catalogue yet, FreqBlog ingests it on demand and returns computed features instead of a miss.
- Richer per-track output. 42 fields per track, including musical key expressed as name, Camelot and Open Key — ready for harmonic mixing, not just the core float set.
- Set-building, not just recommendation.
/transition,/next-trackand/setlistturn a crate into a beat-matched, key-aware arc. - Agent-ready. A hosted MCP server (
mcp.freqblog.com/mcp) lets LLMs and agents call the catalogue directly. - A named, reachable operator. Run by Steven Birring in Nuneaton, UK, ICO-registered, trading as Stackbase/FreqBlog — so you know exactly who to contact and hold accountable.
FAQ
Do I need a Spotify ID to use FreqBlog?
No. You can resolve a track by name + artist, ISRC, MusicBrainz ID, or Spotify ID. ReccoBeats’ documented lookup keys are a Spotify ID or a ReccoBeats UUID.
What happens when a track isn’t in the catalogue?
FreqBlog backfills the track on demand and returns computed features. ReccoBeats’ audio-features endpoint documents a 404 for tracks it can’t find, and offers a separate upload-based extraction endpoint if you already have the audio file.
Can an LLM or agent call these APIs directly?
FreqBlog publishes a hosted MCP server at mcp.freqblog.com/mcp for LLMs and agents. ReccoBeats does not document a hosted MCP server on its site.
Is FreqBlog free?
FreqBlog has a free tier of 1,000 requests per month with no card required, then roughly £0.17 per 1,000 requests. ReccoBeats states it offers completely free API access, subject to internal rate limits — so if cost is your only criterion, ReccoBeats is worth a look.