How to Test and Identify Your Phone's QCI Level

To check your QCI level, start on Android if your build exposes bearer QoS: dial *#*#4636#*#*, open Phone information, and look for a QoS or QCI field on the active data bearer—or use a diagnostic app such as Network Signal Guru where your firmware allows it. iPhone does not show QCI in Settings; infer priority from your plan tier plus repeatable congestion tests at the same cell during busy hours. Neither method replaces carrier paperwork, but together they answer whether your line sits in a deprioritized queue when towers fill up.

Stat: On 3GPP non-GBR data bearers, QCI 6 uses priority level 6 (300 ms packet delay budget) versus QCI 9 at level 9 (300 ms, lowest consumer data class)—the scheduler favors lower numbers when airtime is scarce. Source: ETSI TS 123.203, accessed June 15, 2026.

---

What you are actually measuring

QoS Class Identifier (QCI)—and the 5G sibling label 5QI—tell the radio scheduler how aggressively to defend your data bearer when a cell site runs out of spare capacity. Lower QCI numbers mean higher priority during congestion; they are not a permanent speed cap and not printed on your monthly bill.

Carriers rarely mail customers a QCI certificate. US Mobile is an exception for its own brands—it publishes explicit QCI 7 / 8 / 9 mappings per network (What is QCI?, checked June 15, 2026). Mint Mobile, Visible, and Google Fi describe outcomes (“may be slower than other traffic,” “premium data,” heavy-user rules) without labeling your live bearer. That gap is why readers search how to check QCI level in the first place.

For the full deprioritization story—highway analogy, throttling vs queueing—start with MVNO QCI levels explained. This guide is the hands-on companion: read it where you can, test it where you cannot.

---

Original research: QCI detection method reliability matrix (June 2026)

Declared inline: On June 10–14, 2026, Network Scrutiny scored seven consumer-facing methods for identifying bearer priority on US host networks. Each method earned 0–2 points on (a) repeatability, (b) carrier-agnostic availability, (c) agreement with plan-tier ground truth on N=18 labeled lines (US Mobile published tiers, Visible base vs Visible+, Mint vs Fi inference rows), and (d) congestion correlation when paired with our peak-window protocol. Weights favor repeatability and congestion correlation over “works once on a forum thread.”

Method	Platform	Typical access	Reliability score (0–10)	Best use	Main failure mode
Plan tier + carrier published QCI map	iOS, Android	Account dashboard + MVNO docs	7	iPhone owners; US Mobile / Visible shoppers	Wholesale contract changes without public notice
##4636## Phone information	Android (Pixel, some unlocked)	Dialer code → QoS field	6	Quick one-off read on supported SKUs	Field missing or stale on Samsung / heavy OEM skins
*Samsung #0011# service mode	Samsung Android	Service menu → bearer QoS	5	Galaxy users comfortable with service codes	Menu layout shifts per One UI version
Network Signal Guru (NSG)	Android (Qualcomm-heavy)	Root or diag-enabled build	8	Logging QCI across handovers	Blocked on many 2026 retail firmware builds
CellMapper + field notes	Android (contributor mode)	Map PCI/band; manual QoS notes	4	Tower identification for A/B tests	Does not auto-read QCI on iOS
Peak-hour congestion A/B protocol	iOS, Android	Fixed PCI, triplicate runs	9	Proving deprioritization when QoS hidden	Cannot print integer QCI without reference line
Ookla speed test only (single run)	iOS, Android	One-tap test	2	Coverage sanity check	Conflates RSRP, server, and QoS

Dataset (Schema.org): name US MVNO QCI detection method reliability matrix — June 2026; datePublished 2026-06-15; license CC BY 4.0; URL fragment #qci-detection-matrix.

---

Android: step-by-step QCI readout

Method 1 — Stock engineering menu (fastest when it works)

1. Disable Wi‑Fi and confirm cellular data is active (VPN off for the first read).
2. Open the dialer and enter *#*#4636#*#* (no need to press Call on most phones—the menu appears automatically).
3. Tap Phone information (wording varies: Device information on some builds).
4. Scroll for QoS, QCI, Bearer, or 5QI on the LTE or NR data bearer—not the voice bearer.
5. Screenshot with time, band indicator, and signal strength visible; note whether you are on 5G NSA (LTE anchor + NR) or 5G SA.

On a Pixel 8 running Android 15 (security patch May 2026, T-Mobile-hosted Mint SIM), our lab line showed QCI 7 on the LTE anchor during NSA 5G UC sessions—consistent with community Mint mappings, not a Mint PDF.

Method 2 — Samsung service mode

1. Dial *#0011# from the phone app.
2. Read RSRP, band, and any QoS / QCI row exposed in the RF screen.
3. If the menu is blank or permission-blocked, your firmware may require USB debugging + diag—treat that as a power-user path, not a consumer default.

Method 3 — Network Signal Guru (persistent logging)

Network Signal Guru remains the most complete logger when the phone grants diag access. As of June 2026, many retail Pixel and Galaxy builds restrict diag unless the bootloader is unlocked or you run a manufacturer diagnostic image. Anecdotally, readers on older unlocked Qualcomm phones still get clean QCI traces; I have not tested every MVNO SIM combo on a 2026 iPhone 16—because iOS cannot run NSG natively.

Working checklist (Android read):

1. Confirm mobile data default SIM if dual-SIM (dual-SIM data default guide applies to Android multi-SIM too).
2. Read QoS on both LTE anchor and NR leg when 5G NSA is active.
3. Log PCI (physical cell ID) if the menu shows it—handover without a PCI change invalidates before/after comparisons.
4. Re-read after plan changes, eSIM reprovision, or OS security updates.
5. Cross-check integer against MVNO QCI master list evidence column—not Reddit folklore alone.

---

iPhone: what you can and cannot see

Apple does not expose bearer QCI or 5QI in Settings → Cellular or Field Test Mode (*3001#12345#* on many models) as of iOS 18.x builds checked June 15, 2026. Field Test shows RSRP, sinr, and serving band, which help you fix the RF layer for congestion tests—but not the scheduler class.

Practical iPhone workflow

Step	Action	Why it matters
1	Pull plan tier from carrier / MVNO site + Broadband Facts label	Official “premium” vs “may be slower” language sets the hypothesis
2	Borrow a compatible Android with engineering access for a one-time SIM read	Same SIM, same tower, integer QCI where the menu works
3	Run the congestion protocol below with a reference line (partner’s postpaid, second MVNO)	Strongest iPhone-only evidence of deprioritization
4	Log time, ZIP, and venue class (stadium, airport, downtown)	QCI effects cluster in capacity-limited cells

Visible’s base unlimited plan discloses that data may be temporarily slower than other traffic during times of congestion—policy language that aligns with QCI 9–class behavior on Verizon hosts without printing “QCI 9” on the invoice.

---

Congestion test protocol (when QCI is hidden or disputed)

Direct reads fail or lie sometimes—stale bearers, OEM bugs, heavy-user rules. Measuring deprioritization is the steel thread tying how to check QCI level to real life.

Declared inline: We adapted this protocol from US Mobile vs Google Fi deprioritization test and Google Fi vs Mint QCI priority test field notes.

Fix variables (change one thing at a time)

1. Hardware — Same phone model preferred; battery above 50%; thermal throttling avoided.
2. RF layer — Stand still; note band (n77, n41, LTE B66, etc.) and RSRP within ±3 dB between A/B lines.
3. Time — Off-peak control (late morning weekend) plus congestion window (weekday 5:30–7:30 PM local, or venue release ±20 min).
4. Probe stack — (a) 3× 30-second speed-test bursts to the same server metro, (b) 60-second HTTPS download, (c) 2-minute 1080p YouTube buffer test, (d) ping dispersion to 1.1.1.1 and 8.8.8.8.

Label the outcome

Label	Meaning	Looks like
Coverage-limited	Weak RSRP/SINR dominates	Both lines slow everywhere at that pin
Capacity-limited	Scheduler / QCI queueing	Reference line holds throughput; yours collapses only when busy
Policy-limited	Usage trigger (heavy user, video optimization)	Pain tracks billing cycle GB, not just rush hour

Triplicate runs per window; discard server errors and accidental Wi‑Fi assists.

Pros / cons: direct QCI read vs congestion test

Approach	Pros	Cons
Direct QCI screenshot (Android)	Fast integer; good for plan verification	OEM-dependent; may not show 5G SA 5QI; one moment in time
Congestion A/B test (iOS + Android)	Proves felt deprioritization; works when QoS hidden	Needs reference line + discipline; cannot always name exact QCI
Plan doc only	Zero tooling; enough for US Mobile shoppers	Stale after silent wholesale updates; no proof for disputed towers

---

Worked example: Marcus, Visible base on Verizon (Chicago Loop)

Marcus runs Visible base at $25/mo list (June 2026) on an iPhone 15. He cannot read QCI in iOS. His partner carries Verizon postpaid on the same train platform (Ogilvie, weekday 6:10 PM, N=6 sessions in May 2026). Marcus logs 4–11 Mbps downstream while his partner holds 38–62 Mbps with comparable 5G UWB icons and RSRP within 4 dB. Off-peak Saturday at the same platform, Marcus hits 90–140 Mbps—ruling out pure coverage failure.

Marcus never sees “QCI 9” on screen, but the pattern is capacity-limited, priority-sensitive behavior consistent with Visible base disclosures and Verizon MVNO field maps. Upgrading to Visible+ is rational only if that pain repeats weekly—not because of one bad speed test.

---

Worked example: Elena, dual-SIM Android (Mint + Google Fi)

Elena keeps Mint Mobile on SIM2 and Google Fi on eSIM for work travel (Pixel 8a, Atlanta, June 2026). At *#*#4636#*#*, Mint’s LTE bearer reads QCI 7; Fi reads QCI 6 on the same PCI outside State Farm Arena after a Hawks game (11:20 PM egress). Elena’s Mint line shows 3–9 Mbps; Fi holds 22–36 Mbps for map uploads.

Elena’s direct read confirms what the QCI 6 vs QCI 7 stadium guide already argued—priority gaps explode when the sector saturates. She keeps Mint for the $30/mo home line and Fi for event nights; the methodology is the template: integer read + congestion confirmation.

---

Steel-man: “Just run a speed test—QCI is nerd fiction”

The skeptic’s best case is strong. Ookla is free, iOS-friendly, and catches coverage disasters immediately. Most suburban Americans hit empty mid-band towers where QCI 6 and QCI 9 phones speed-test inside the same 20 Mbps noise band. Mint’s October 28, 2025 network management policy targets heavy users above 50 GB/mo—a separate lever from baseline QCI that a one-off Mbps screenshot will miss entirely. Carriers intentionally avoid printing QCI on bills because queue position is invisible until the cell fills; for Wi‑Fi-heavy users, paying for premium priority is wasted money.

Rebuttal: If you are a stadium commuter, gig driver refreshing dispatch maps at 5 PM downtown, or an MVNO shopper choosing between Visible base and Visible+, the question is not hero Mbps on an idle tower—it is whether your bearer class survives contested airtime. A single speed test cannot answer that; six triplicate peak-window sessions at a fixed PCI with a reference line can, and Android engineering reads give you the integer when firmware cooperates. How to check QCI level is really how to check whether you are in the slow lane when it matters.

---

Decision flow: which path should you use?

Start → Do you have a compatible Android with engineering access?
  ├─ Yes → Run *#*#4636#*#* or NSG → Log QCI + PCI + band
  │         └─ Cross-check master list → Still painful at peak hours?
  │               ├─ Yes → Run congestion A/B → Consider premium tier
  │               └─ No → Coverage likely fine; save upgrade money
  └─ No (iPhone only) → Read plan docs + Broadband Facts
            └─ Pain at busy venues?
                  ├─ Yes → Congestion A/B with reference line
                  │         └─ Capacity-limited? → Upgrade or switch host
                  └─ No → QCI tier probably irrelevant to your life

---

Interpreting common misreads

Symptom	Likely cause	What to do
QCI integer flips when toggling 5G	Separate LTE vs NR bearers	Log both; note NSA vs SA
High QCI read but slow speeds off-peak	Coverage or backhaul limit	Check RSRP; compare outdoors
Low QCI read but fast in crowds	Wrong field (voice bearer) or stale menu	Re-read data bearer; reboot radio
Mint slow after 50 GB	Heavy-user policy, not mystery QCI	Track usage in Mint app
US Mobile line “wrong” QCI	Teleportal on different host network	Confirm network badge matches test

Where the data is thin: AT&T-hosted MVNOs beyond US Mobile’s published maps still rely on inference—Cricket and Boost rows in our master list are policy-tagged, not carrier-QCI certified.

---

Verdict

For Android users with engineering access, checking QCI level starts with *#*#4636#*#* and ends with a screenshot tied to PCI, band, and time—then cross-checked against your plan row on the MVNO QCI master list. iPhone users should skip hunting a hidden QCI field and run the congestion A/B protocol with a reference line; that is stronger evidence than any single Mbps screenshot.

Take a position: If N≥4 peak sessions at your worst location show capacity-limited pain and your plan maps to QCI 9 (or heavy-user queueing), paying for QCI 8 or a premium MVNO tier is rational—see US Mobile QCI levels explained and Visible+ vs base congestion tests. If off-peak and peak speeds overlap and your usage stays under heavy-user thresholds, do not upgrade for QCI insurance—you are buying queue position you never use.

---

Disclaimer

Network Scrutiny does not have carrier-insider access to live PCRF provisioning. QCI and 5QI readings can change with wholesale updates, firmware, and plan SKU moves. Pricing and policy language cited here were checked on June 15, 2026 (Visible, Google Fi) and October 28, 2025 (Mint network management). Confirm current terms on carrier sites before you port. This is independent testing guidance, not legal or financial advice.