CekuraBench
Book a demo

Phonic-1

Every modelCompare

1030

ELO

5th of 12

31.4%

Pass³

6th of 12

1.99s

Latency

4th of 12

ELO Rating vs. Latency P50

Higher wins more head-to-head calls (1000 = average); further left replies faster

Best quadrant
ELO Rating
Median latency (s)

Phonic-1 Against Every Other Model

Each model took the same recorded human callers on live calls. Pick a measure to see where Phonic-1 ranks.

Rating from head-to-head calls: on every call each pair of models is compared, the better call wins. 1000 is the average model. Higher is better.

#ModelELO Rating
  1. 01Grok Voice 3.01154
  2. 02GPT-Live 1 + GPT-6 Sol1069
  3. 03Gemini 3.1 Flash Live Preview1062
  4. 04GPT Realtime 2.11054
  5. 05Grok Voice Think Fast 2.01030
  6. 06Phonic-11030
  7. 07Azure Realtime1017
  8. 08Gemini 3.8 Live Extended Thinking1016
  9. 09Gemini 3.8 Live968
  10. 10GPT Realtime 2.1 Mini921
  11. 11Deepslate Opal847
  12. 12Amazon Nova 2 Sonic832
  • refCascade955

Where Phonic-1 Is Strong and Weak

Pick a skill family. The shape is this model on every skill in it; the dashed line is the median of the other models.

Phonic-1Median of the other models
67Hearing Score5th of 12+1 vs median
SkillScore · vs median
  • Prosodic Cue Understanding87%+14
  • Waiting Through Hold91%+7
  • Emotion Recognition90%+6
  • Side-Speech Rejection98%+3
  • Multi-Speaker Handling44%+2
  • Low-Confidence Confirmation20%+1
  • Degraded-Channel Robustness75%0
  • Entity Extraction Accuracy32%−6

Share of calls where the skill was needed and handled, strongest first. Hover a point on the chart for both values.

Example Calls from Phonic-1

Banking and healthcare callers. Each shows what the caller needed, what made the call hard, and how the model handled it.

First prenatal visit

Healthcare
Upset callerTalks over the agentNeeds consentCancels

Finished the task

Handled well: Emotion Recognition, Overlapping-Speech Recovery, Consent Verification

Struggled with: Cancellation Handling

Listen to the call

Urgent visit for a feverish child

Healthcare
Background voicesSlow systemAsks to holdAsks for a number

Finished the task

Handled well: Side-Speech Rejection, Tool-Latency Handling, Mid-Utterance Pause Tolerance

Listen to the call

Reschedule a specialist visit

Healthcare
Bad connectionSeveral peopleTalks over the agentChanges their mind

Finished 5 of 7 steps

Handled well: Multi-Speaker Handling, Overlapping-Speech Recovery, Correction Handling

Struggled with: Degraded-Channel Robustness

Listen to the call

Update insurance details

Healthcare
Spells names and numbersBad connectionInterruptsNeeds consent

Finished the task

Handled well: Degraded-Channel Robustness, Backchannel vs Barge-in Detection, Consent Verification

Struggled with: Entity Extraction Accuracy

Listen to the call

Loan payoff quote

Banking
Noisy lineHesitant callerAsks to holdNeeds consent

Finished 5 of 7 steps

Handled well: Prosodic Cue Understanding, Mid-Utterance Pause Tolerance, Consent Verification

Struggled with: Low-Confidence Confirmation

Listen to the call

Change a mortgage payment date

Banking
Hold musicInterruptsNeeds a personNeeds consent

Finished 5 of 7 steps

Handled well: Waiting Through Hold, Backchannel vs Barge-in Detection, Human Handoff

Listen to the call

Business loan documents

Banking
Slow systemAsks to holdAsks for a number

Finished the task

Handled well: Tool-Latency Handling, Mid-Utterance Pause Tolerance, Redundant Re-prompt Avoidance

Struggled with: Premature-Response Avoidance

Listen to the call

Request a medical records transfer

Healthcare
Slow systemNeeds consentAsks for a number

Finished 2 of 6 steps

Handled well: Tool-Latency Handling, Redundant Re-prompt Avoidance, Numeric Read-back

Struggled with: Consent Verification, Hallucinated-Success Avoidance

Listen to the call

Payment plan for missed loan payments

Banking
Hesitant callerSeveral peoplePauses mid-sentenceNeeds consent

Finished 2 of 5 steps

Handled well: Prosodic Cue Understanding, Endpointing, Premature-Response Avoidance

Struggled with: Multi-Speaker Handling

Listen to the call

Remove a sold car from a policy

Banking
Noisy lineBackground voicesInterruptsChanges their mind

Finished the task

Handled well: Side-Speech Rejection, Backchannel vs Barge-in Detection, Correction Handling

Struggled with: Low-Confidence Confirmation, Consent Verification

Listen to the call

Latency at P50, P90 and P95

Phonic-1 starts replying in 1.99s on a typical turn and 2.40s on the slowest turn in twenty, 4th fastest of 12 at the median.

ModelP50P90P95
  • Cascade
    2.06s2.49s2.60s
Seconds from the caller finishing to the agent replying. The bar is the typical turn, P50.

Phonic-1 Head to Head

Every comparisonAll 12 models

How Much Audio Phonic-1 Was Tested On

210 live calls: 70 scenarios, each run three times. The median call ran 4:23.

Live calls
210
Scenarios
70
Banking calls
105
Healthcare calls
105
Replies per call
18.3
Median call
4:23

70 scenarios, the same for every model.

Phonic-1 FAQ

  • Phonic-1 has an ELO rating of 1030, 5th of 12 ranked models. It passed 22 of 70 callers on all three runs, a Pass³ of 31.4%.

  • Phonic-1 starts replying 1.99s after the caller finishes at P50, 2.32s at P90 and 2.40s at P95: 4th fastest of 12 at P50, measured on the call audio.

  • Phonic-1 falls furthest behind on Language Switching: 32%, against a median of 67%.

  • Phonic-1 has a task completion rate of 24.8%, 5th of 12. A call counts only when every step the caller needed was done.

  • There is no verified per-minute rate for Phonic-1, so it is left out of the cost ranking.

  • Cekura ran Phonic-1 as the whole agent with the same prompt, tools and phone connection as every other model. It took the same 70 recorded real human callers on live calls, three times each (210 calls), and each call was graded on 25 skills across Hearing, Turn Taking, Task Handling and Speaking.

Test your voice agent the same way.

Cekura runs these simulated calls on your agent and scores every one.

Book a demoStart free