Comparison

LibreQoS and Preseem

These are the two closest approaches among QoE platforms: both reject deep packet inspection, and both use active queue management and fair queueing rather than application classification. Preseem shapes at the subscriber plan and access point levels. LibreQoS nests every circuit inside a shaping hierarchy that also covers sites, OLTs, and backhaul links. Preseem capabilities below are stated in their own published datasheets and materials.

Closest, Not Identical

Both control queues. The question is where, and how finely.

Preseem AQM datasheet describes FQ-CoDel with bulk and interactive flow separation, plan-rate enforcement, and a deliberately anti-DPI stance — a philosophy LibreQoS shares. Its Automatic AP Capacity Management datasheet adds automatic bandwidth control on busy access points, included in all tiers. Preseem deserves credit here: FQ-CoDel is published AQM from the same research family, which is exactly why this comparison stays technical.

Preseem enforces at the levels its docs describe: QoE-optimized plan enforcement per subscriber and AQM capacity management per access point — the TwoLevelMQ policy — with manual site bandwidth caps available. Those levels matter, and Preseem shapes them with published AQM. LibreQoS adds the dimension above them: every subscriber circuit runs CAKE per-flow queueing inside an HTB envelope, and every circuit nests under explicit queue classes for its AP, site, OLT, and backhaul. When a backhaul link congests, the hierarchy shapes that contention at the link instead of letting it degrade every subscriber upstream.

Beyond enforcement, the platforms optimize for different jobs. Preseem is a network intelligence platform: one view across vendors, support triage, churn risk by site, revenue reporting, and BEAD/EA-CAM compliance data. LibreQoS stays closer to the network: RTT, retransmit, QoO, and flow evidence per circuit, with Insight for multi-node history and ASN analysis. Business reporting is left to the billing stack.

Comparison

Platform by platform, with sources

Primary role
LibreQoS
Inline queue control across the full topology, with analysis built on the traffic it shapes LibreQoS docs
Preseem
Network intelligence platform with AQM-based capacity management Preseem docs
Queueing algorithm
LibreQoS
CAKE (or fq_codel) leaf qdiscs providing per-flow queues inside per-circuit HTB envelopes LibreQoS docs
Preseem
FQ-CoDel with bulk and interactive flow separation Preseem AQM datasheet
Scope of enforcement
LibreQoS
Every subscriber circuit, plus site, AP, OLT, and backhaul capacity as explicit queue classes LibreQoS docs
Preseem
QoE-optimized plan enforcement per subscriber and automatic AQM capacity management per access point, with manual site bandwidth caps Preseem AP capacity docs
Topology hierarchy
LibreQoS
Arbitrary-depth hierarchy: circuits nest under AP, site, OLT, and backhaul classes, so congestion on a shared link is shaped at that link LibreQoS docs
Preseem
Two enforcement levels — subscriber plan and access point (TwoLevelMQ); no published shaping hierarchy for links above the access layer Preseem AP capacity docs
Protocol and DPI stance
LibreQoS
Protocol-agnostic by construction: 5-tuple flow classification, no payload inspection LibreQoS docs
Preseem
Vendor- and application-agnostic; positions AQM explicitly against DPI Preseem AQM datasheet
Where visibility comes from
LibreQoS
The shaper itself: RTT, retransmits, flows, and QoO observed on the traffic it handles, plus Insight for multi-node history and ASN analysis Insight
Preseem
Integrations and existing network gear, normalized into a single view across vendors Preseem docs
Support workflow
LibreQoS
From subscriber symptom to circuit evidence to the likely cause, in one operator workflow Insight
Preseem
Triage from one screen, with root-cause layering for network, AP/OLT, or in-home Preseem docs
Business intelligence
LibreQoS
Focuses on network evidence; business reporting lives in your billing stack LibreQoS docs
Preseem
Churn risk by site, revenue by site and equipment, upsell candidates, BEAD/EA-CAM reporting Preseem positioning
Access technology focus
LibreQoS
Wireless, fiber, and dense shared-capacity networks, with one shaping model across them LibreQoS docs
Preseem
Fixed wireless, fiber, DSL, and cable tiers, normalized into one interface Preseem positioning
Licensing model
LibreQoS
Subscriber-based subscription. No throughput licensing on the shaping core LibreQoS pricing
Preseem
Subscriber-based subscription for the analytics platform Preseem positioning
Independent verification
LibreQoS
Public Internet Quality Test anyone can run, plus observe mode for honest before/after baselines Quality test
Preseem
Vendor-stated measurable results within 30 days, with a 30-day trial Preseem positioning

Preseem capabilities are summarized from Preseem's published website, datasheets, and help center. LibreQoS capabilities are summarized from the LibreQoS documentation and pricing page.

Ask Both Vendors

Questions worth asking in any Preseem evaluation

  1. 01When an access point is congested, how finely is the relief applied — at the AP, per plan rate, or per subscriber flow?
  2. 02When the backhaul itself congests, is it shaped as part of the enforcement hierarchy, or only monitored?
  3. 03Which metrics come from our own gear via integrations, and which are measured by the platform itself?
  4. 04How is latency under load measured — idle probes, or actual subscriber traffic during peak?
  5. 05What does the enforcement story look like on fiber, where integrations rather than appliances provide the data?
  6. 06Can we run a before/after quality test on our own network during evaluation?

Verify It Yourself

Don't decide on vendor claims

Run the public Internet Quality Test on a busy-hour circuit, deploy LibreQoS on spare hardware or a pilot site, and run it again. Idle and loaded latency, loss, throughput, and a QoO grade — measured on your own network.

Keep Comparing

Other comparisons

Deploy

Evaluate LibreQoS against your requirements

Run the 1-line install on a pilot site, point your topology at it, and measure busy-hour latency before and after.