Softwr

Databases · head to head

turbopuffer vs VerneMQ

turbopuffer logo

turbopuffer

Databases

Closed-source vector and full-text search service built directly on object storage, with cold queries measured in seconds rather than milliseconds.

From
$16/month
Rated
-
Full reviewWebsite
VerneMQ logo

VerneMQ

Databases

Erlang MQTT broker whose source is Apache 2.0 but whose official binaries need a paid subscription

From
Free
Rated
-
Full reviewWebsite

The short version

  • Only VerneMQ has a free tier, so it costs nothing to try first.
  • Each has a real cost: turbopuffer a cold namespace pays object storage latency on the first query, with a documented p90 around 1,214 ms on a million documents, so any interactive search box needs the data kept warm or the user waits about a second.; VerneMQ the official binaries and Docker images are not Apache 2.0 but sit under a EULA requiring a yearly commercial subscription, a distinction easy to miss and awkward to discover during a licence audit.
  • They diverge on capability: turbopuffer covers Object storage architecture, VerneMQ covers Erlang/OTP clustering.
  • Prices and features above were last checked on 31 August 2026.

Where they differ

Only the attributes on which turbopuffer and VerneMQ actually diverge.

Attributes where turbopuffer and VerneMQ differ
AttributeturbopufferVerneMQ
Starting price$16/monthFree
Pricing modelsubscriptionquote
Free tierNoYes
PlatformsWebLinux, Docker, macOS, Kubernetes

Identical on both: user rating (Not yet rated), category (Databases).

What each one covers

Drawn from each product's published feature list. An absence here means we hold no record of it - not that the product lacks it.

Only in turbopuffer

  • Object storage architecture
  • Namespaces
  • Vector search
  • Full-text search
  • Attribute filtering
  • Documented limits
  • Configurable consistency
  • Durable writes

Only in VerneMQ

  • Erlang/OTP clustering
  • MQTT 5.0 support
  • Plugin system
  • Backpressure handling
  • Bridge support
  • Metrics export
  • MQTT over WebSockets
  • Pluggable auth backends

What people use each for

The jobs each tool is most often brought in to do.

turbopuffer

  • A product with one search index per customer and thousands of customers, most of whose data is idle on any given daynot VerneMQ
  • Very large corpora where holding every vector in memory is the dominant cost and occasional cold-query latency is acceptablenot VerneMQ
  • Hybrid retrieval combining BM25 and vector search where running and synchronising two separate systems is the problem being solvednot VerneMQ
  • Retrieval for agent and assistant products where indexes are created and destroyed frequently and per-index overhead must be near zeronot VerneMQ

VerneMQ

  • An industrial operator that wants an MQTT broker with predictable memory behaviour and no data integration features it will not usenot turbopuffer
  • A team building from source to stay strictly under Apache 2.0 terms with no vendor licence entanglementnot turbopuffer
  • A deployment needing custom authentication logic implemented as a plugin in Lua or over a webhooknot turbopuffer
  • An organisation that wants a broker maintained by a small European company rather than by a vendor that keeps changing licencesnot turbopuffer

Where each one falls short

Documented limitations, not opinions. Every one is a constraint you would hit in normal use.

turbopuffer

  • A cold namespace pays object storage latency on the first query, with a documented p90 around 1,214 ms on a million documents, so any interactive search box needs the data kept warm or the user waits about a second.
  • Queries are eventually consistent by default, and after roughly 128 MiB of outstanding writes new data is invisible until indexed, which the vendor puts at tens of seconds for small namespaces and tens of minutes for large ones, so a bulk re-index is not immediately queryable.
  • It is closed source with no community edition, so single-tenant or bring-your-own-cloud deployment is a commercial negotiation rather than a deployment choice, and there is no path to running it yourself if the relationship ends.
  • Per-namespace ceilings, roughly 10,000 writes per second, 32 MB/s and 500 million documents per shard, mean a single enormous index has to be sharded across namespaces by your application rather than by the service.
  • It is a search engine, not a database: there are no joins, no cross-document transactions and no SQL, so it sits beside a primary datastore and keeping the two in step is work that belongs to you.

VerneMQ

  • The official binaries and Docker images are not Apache 2.0 but sit under a EULA requiring a yearly commercial subscription, a distinction easy to miss and awkward to discover during a licence audit.
  • Octavo Labs is a very small company, so support depth, response times and the bus factor on the codebase are materially thinner than at HiveMQ or EMQ.
  • There is no data integration or rule engine layer, so routing messages into a database means writing and operating your own consumer service.
  • Operating an Erlang cluster requires runtime knowledge that most teams do not have and will use for nothing else in their stack.
  • There is no vendor-managed cloud offering, so every deployment is self-operated with the infrastructure and on-call cost that implies.

Pricing, plan by plan

turbopuffer

$16/month
  • Launch$16/month
    • All database features
    • Multi-tenancy deployment
    • SOC2 & GDPR-ready DPA
  • Scale$256/month
    • Everything in Launch
    • HIPAA-ready BAA
    • Single Sign-On (SSO)
  • Enterprise$4096/month
    • Everything in Scale
    • Single-tenancy & BYOC deployment options
    • Private networking

VerneMQ

Free
  • Source buildFree
    • Apache 2.0 licensed source from GitHub
    • Full clustering and plugin capability
    • You compile and package it yourself
  • Binary packages and Docker images$undefined/year
    • Covered by the VerneMQ EULA, not Apache 2.0
    • Yearly usage subscription expected for commercial use
    • Official builds and Docker images
  • Commercial support$undefined/year
    • Evaluation, customisation and operations assistance
    • Custom development
    • Long-term maintenance agreements

Which should you pick?

Choose turbopuffer if

  • You need object storage architecture.
  • You also want namespaces.
turbopuffer review

Choose VerneMQ if

  • You need erlang/otp clustering.
  • You want to start without paying.
  • You work on Linux, Docker, macOS, Kubernetes.
  • You also want mqtt 5.0 support.
VerneMQ review

Questions people ask

Is turbopuffer or VerneMQ better?
Neither clearly leads. turbopuffer starts at $16/month and VerneMQ at Free, and user ratings are close enough to be indistinguishable. Choose on capability and platform support.
Which is cheaper, turbopuffer or VerneMQ?
VerneMQ has a free tier; the other does not. Paid plans start at $16/month for turbopuffer and Free for VerneMQ.
Does turbopuffer or VerneMQ run on more platforms?
turbopuffer runs on Web. VerneMQ runs on Linux, Docker, macOS, Kubernetes.
Can I use VerneMQ for free?
Yes. VerneMQ has a free tier, so you can try it without paying. turbopuffer starts at $16/month.
What is turbopuffer best used for?
turbopuffer is most often used for a product with one search index per customer and thousands of customers, most of whose data is idle on any given day, very large corpora where holding every vector in memory is the dominant cost and occasional cold-query latency is acceptable, hybrid retrieval combining bm25 and vector search where running and synchronising two separate systems is the problem being solved, retrieval for agent and assistant products where indexes are created and destroyed frequently and per-index overhead must be near zero. Of those, a product with one search index per customer and thousands of customers, most of whose data is idle on any given day and very large corpora where holding every vector in memory is the dominant cost and occasional cold-query latency is acceptable are not what VerneMQ is typically brought in for.
What can turbopuffer do that VerneMQ cannot?
turbopuffer covers Object storage architecture, Namespaces, Vector search, Full-text search. VerneMQ covers Erlang/OTP clustering, MQTT 5.0 support, Plugin system, Backpressure handling.

Answered from the vendors’ own pages

turbopuffer: Can I self-host turbopuffer?

There is no open source or community edition. Single-tenant and bring-your-own-cloud deployments exist as commercial arrangements, but there is no way to run it independently of the vendor.

VerneMQ: Is VerneMQ free?

The source is Apache 2.0 and free. The official binary packages and Docker images are covered by a separate EULA that expects a yearly fee for commercial use.

turbopuffer: How fast is it really?

Warm queries perform comparably to in-memory search engines. Cold queries, where data is not cached, have a documented p90 around 1,214 ms on a million documents. Write p90 is around 248 ms for a 512 KB upsert because writes go straight to object storage.

VerneMQ: Is the project still maintained?

Yes. Octavo Labs AG in Zurich continues to publish 2.x releases, most recently in 2026.

turbopuffer: Is it consistent?

Eventually consistent by default, with the vendor reporting that over 99.8% of queries return consistent data. Strong consistency can be requested per query at a latency cost. Large write bursts have a longer visibility delay while indexing catches up.

VerneMQ: Does it have a managed cloud?

No. Every deployment is self-hosted, with commercial support available from Octavo Labs.

turbopuffer: What is it best at?

Large numbers of namespaces where most are idle. The architecture makes cold data cheap to keep, which is exactly the shape of a multi-tenant product with a long tail of inactive customers.

VerneMQ: How does it compare to EMQX?

Narrower in features and without a rule engine, but with a simpler licence story for source builds after EMQX moved to BSL.

turbopuffer: What are the hard limits?

Up to 128 billion documents and 256 TB per namespace, 500 million documents per shard, 64 MiB per document, 10,752 dense vector dimensions, roughly 10,000 writes per second per namespace and a maximum result set of 10,000.

Share

Related pages

More on turbopuffer

More on VerneMQ

Keep looking

Other head to heads