CekuraBench
Book a demo

Grok Voice 3.0

Every modelCompare

1154

ELO

1st of 12

42.9%

Pass³

1st of 12

1.60s

Latency

2nd of 12

$0.080

Cost/Min

5th of 8

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)

Grok Voice 3.0 Against Every Other Model

Each model took the same recorded human callers on live calls. Pick a measure to see where Grok Voice 3.0 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 Grok Voice 3.0 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.

Grok Voice 3.0Median of the other models
75Hearing Score1st of 12+9 vs median
SkillScore · vs median
  • Degraded-Channel Robustness96%+21
  • Entity Extraction Accuracy54%+21
  • Prosodic Cue Understanding93%+21
  • Waiting Through Hold92%+8
  • Low-Confidence Confirmation27%+8
  • Emotion Recognition91%+6
  • Side-Speech Rejection100%+5
  • Multi-Speaker Handling47%+4

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

Example Calls from Grok Voice 3.0

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 agentSlow systemNeeds a person

Finished the task

Handled well: Emotion Recognition, Overlapping-Speech Recovery, Tool-Latency Handling

Listen to the call

Urgent visit for a feverish child

Healthcare
Background voicesSlow systemAsks to holdNeeds consent

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 the task

Handled well: Degraded-Channel Robustness, Multi-Speaker Handling, Overlapping-Speech Recovery

Listen to the call

Update insurance details

Healthcare
Spells names and numbersBad connectionInterruptsNeeds consent

Finished the task

Handled well: Entity Extraction Accuracy, Degraded-Channel Robustness, Backchannel vs Barge-in Detection

Listen to the call

Loan payoff quote

Banking
Noisy lineHesitant callerAsks to holdNeeds consent

Finished 6 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 musicSeveral peopleInterruptsNeeds a person

Finished 4 of 6 steps

Handled well: Waiting Through Hold, Multi-Speaker Handling, Backchannel vs Barge-in Detection

Listen to the call

Business loan documents

Banking
Slow systemAsks to holdNeeds consentAsks for a number

Finished 2 of 3 steps

Handled well: Tool-Latency Handling, Mid-Utterance Pause Tolerance, Consent Verification

Struggled with: Premature-Response Avoidance

Listen to the call

Request a medical records transfer

Healthcare
Slow systemNeeds consentCancelsAsks for a number

Finished 3 of 5 steps

Handled well: Tool-Latency Handling, Consent Verification, Redundant Re-prompt Avoidance

Struggled with: Cancellation Handling

Listen to the call

Payment plan for missed loan payments

Banking
Hesitant callerSeveral peoplePauses mid-sentenceNeeds consent

Finished 4 of 6 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 lineChanges their mindSystem errorAsks for a number

Finished 5 of 8 steps

Handled well: Correction Handling, Tool Error Handling, Redundant Re-prompt Avoidance

Struggled with: Low-Confidence Confirmation

Listen to the call

Latency at P50, P90 and P95

Grok Voice 3.0 starts replying in 1.60s on a typical turn and 2.20s on the slowest turn in twenty, 2nd 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.

Grok Voice 3.0 Head to Head

Every comparisonAll 12 models

How Much Audio Grok Voice 3.0 Was Tested On

210 live calls: 70 scenarios, each run three times. The median call ran 3:38.

Live calls
210
Scenarios
70
Banking calls
105
Healthcare calls
105
Replies per call
16.1
Median call
3:38

70 scenarios, the same for every model.

Grok Voice 3.0 FAQ

  • Grok Voice 3.0 has an ELO rating of 1154, 1st of 12 ranked models. It passed 30 of 70 callers on all three runs, a Pass³ of 42.9%.

  • Grok Voice 3.0 starts replying 1.60s after the caller finishes at P50, 1.94s at P90 and 2.20s at P95: 2nd fastest of 12 at P50, measured on the call audio.

  • Grok Voice 3.0 is at or above the median score on every call type.

  • Grok Voice 3.0 has a task completion rate of 34.8%, 2nd of 12. A call counts only when every step the caller needed was done.

  • $0.080 per minute of call at xAI's published rates, 5th cheapest of 8 priced models. Excludes a text-input charge the vendor lists without a unit.

  • Cekura ran Grok Voice 3.0 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