Komms · Latency benchmark
Voice AI latency, measured on live calls: a Komms reply starts in 150 ms
Komms reduces talk time 30–40% vs. your agents on the same script.
- $0.08 a minute, phone call included
- $15/mo per phone number
On live phone calls, a Komms voice agent hands its first words to the voice a median of 150 milliseconds after the caller's words arrive. 93% of replies start within 200 ms. These numbers come from the call logs of 35 real calls, 371 caller turns, on October 9 and 10, 2026. Below: every number, how we measured it, and exactly what the clock covers.
The numbers
Time is counted per caller turn, from the moment the caller's words arrive as text. All times are in milliseconds (1,000 ms = 1 second).
| What we timed | Median | 90% within | 95% within | Slowest |
|---|---|---|---|---|
| First words handed to the voice (time to first word, TTFW) | 150 ms | 187 ms | 209 ms | 375 ms |
| The full answer handed to the voice | 157 ms | 500 ms | 745 ms | 1,794 ms |
| Greeting ready when the call connects | 89 ms | 141 ms | 173 ms | 324 ms |
- First words
- 93% of turns started within 200 ms. Every turn started within 500 ms.
- Full answer
- 77% of answers were handed to the voice within 200 ms, 90% within 500 ms and 98% within one second.
- Short bridge phrases
- On 17.5% of turns the agent opened with a short phrase such as "let me check" while it worked out the answer. On those turns the full answer followed a median of 513 ms after the caller's words arrived. Those turns are counted in every row above.
Where the time in a call goes
We added up every second of the same 35 calls. A typical call had 11 caller turns.
| Part of the call | Share of call time | Typical call |
|---|---|---|
| Komms deciding what to say | 2.6% | 1.8 seconds in total |
| Komms speaking | 58.8% | about 60 seconds |
| The caller thinking and speaking | 38.6% | about 38 seconds |
Across a whole call, the agent spends under two seconds working out its replies. The rest is the conversation itself.
How we measured
- Real calls
- 35 inbound phone calls on a live benefits-screening campaign, October 9–10, 2026. Each call ran the full script and ended with a transfer to a person. Median call length: about two minutes. No test calls and no lab setup.
- Every turn counted
- All 371 caller turns in those calls are in the numbers, including the slowest. Nothing was trimmed. Across all 360 calls on the campaign in the same two days, the median over 705 turns was 149 ms.
- Where the clock starts
- When the caller's words arrive from speech-to-text as text. Komms writes that moment to the call log.
- Where the clock stops
- When Komms hands the first words of its reply to the text-to-speech voice. Komms writes that moment to the call log too.
- Where the numbers come from
- The Komms call log. Every call report shows the timing of every turn, so any Komms customer can check the same numbers on their own calls.
What time to first word covers
A voice reply has four steps. Time to first word is steps 2 and 3: the part the voice agent platform controls.
| Step | What happens | In our 150 ms? |
|---|---|---|
| 1. Hear | Speech-to-text notices the caller has finished and turns the speech into text. | Before the clock starts |
| 2. Decide | The agent works out what to say. | Yes |
| 3. Hand off | The first words of the reply go to the voice. | Yes |
| 4. Speak | The voice turns the words into sound and the phone network carries it to the caller. | After the clock stops |
What the caller hears is all four steps added together. Steps 1 and 4 add time on every voice platform, so the pause a caller hears is longer than 150 ms.
Three terms that get mixed up. Time to first word (TTFW) is steps 2 and 3. Time to first audio is how long a text-to-speech voice takes to make its first sound, a slice of step 4. End-to-end or voice-to-voice latency is all four steps. A number is only comparable with another number that uses the same clock.
What voice AI platforms publish about their own latency
Each figure below is the vendor's own, read on the vendor's own page on October 10, 2026. Vendors start and stop the clock in different places, so we print what each one says it measures and leave the comparison to you.
Voice agent platforms
| Platform | Published figure | What it says it measures | Source |
|---|---|---|---|
| Komms (KX21) | 150 ms median, 209 ms at the 95th percentile | Caller's words arrive as text → first words handed to the voice. Live phone calls, 371 turns. | This page |
| Twilio ConversationRelay | 491 ms median, 713 ms at the 95th percentile | "Internal benchmarks … using different models." Start and end points not defined on the page. | twilio.com |
| Retell AI | ~600 ms | "~600ms latency." The page does not say median or average. | retellai.com |
| Vapi | under 500 ms average; "sub-600ms response times" | "<500ms average latency" on the home page; "Sub-600ms response times" in the docs. | vapi.ai, docs |
| Synthflow | under 500 ms | "sub-500ms latency." No breakdown given. | synthflow.ai |
| PolyAI | under 300 ms | "<300ms model response, voice to voice." | poly.ai |
| Bland AI | under 200 ms | "End-to-end response runs under 200ms on the standard model tier." | bland.ai |
| Telnyx Voice AI | under 200 ms | "Sub-200ms latency." The page ties this to network round-trip time between its GPUs and phone network. | telnyx.com |
Text-to-speech voices (one step inside a voice agent)
| Voice | Published figure | What it says it measures | Source |
|---|---|---|---|
| ElevenLabs Flash v2.5 | ~75 ms | "Ultra-low latency (~75ms)", footnoted as excluding application and network latency. | elevenlabs.io |
| Cartesia Sonic | under 90 ms | "First audio in under 90ms." | cartesia.ai |
| Deepgram Aura-2 | under 200 ms | "Sub-200ms time-to-first-byte." | deepgram.com |
The Komms row uses the time-to-first-word clock described above. The other rows use each vendor's own clock. Where a vendor says "end-to-end" or "voice to voice", its figure includes steps that our 150 ms leaves out.
Check it yourself
Watch the short version. What is TTFW? (52 seconds)
Run it on your own calls. Komms is $0.08 a minute with the phone call, the AI and a standard voice included, and a phone number is $15 a month. New accounts start with 50,000 free tokens, about 625 minutes, with no credit card. Every call you run shows the same per-turn timing in its call report.
Build your first agent in about 10 minutes → How Komms works Pricing
Questions
- What is time to first word (TTFW) in voice AI?
- Time to first word is how long a voice agent takes to start its reply: from the moment the caller's words arrive as text to the moment the first words of the reply are handed to the voice. It covers the agent deciding what to say. It does not include speech-to-text noticing the caller has finished, or the voice and phone network delivering the sound.
- How fast does a Komms voice agent reply?
- On 371 caller turns from 35 live phone calls on October 9 and 10, 2026, Komms handed its first words to the voice a median of 150 milliseconds after the caller's words arrived. 90% of turns were within 187 ms, 95% within 209 ms, and the slowest was 375 ms.
- Is 150 ms the pause the caller hears?
- No. 150 ms is the time Komms takes to decide and start its reply. The caller also waits for speech-to-text to notice they have finished and for the voice and the phone network to deliver the sound. Those steps add time on every voice platform.
- What is the difference between TTFW, time to first audio and end-to-end latency?
- Time to first word is the agent deciding what to say and handing it to the voice. Time to first audio is how long a text-to-speech voice takes to make its first sound. End-to-end or voice-to-voice latency is everything from the caller finishing to the caller hearing the reply. Only numbers that use the same clock can be compared.
- How much of a call does the AI spend thinking?
- On the 35 live calls we measured, Komms spent 2.6% of the call deciding what to say: about 1.8 seconds in total across a typical 11-turn, two-minute call. Komms speaking took 58.8% and the caller thinking and speaking took 38.6%.
- Can I check these numbers on my own calls?
- Yes. Every Komms call report shows the timing of each turn, taken from the same call log these numbers come from. New accounts start with 50,000 free tokens, about 625 minutes, with no credit card.