LiveKit
- Strength
- Ranks second on repeatable reliability with 99.19% infrastructure-clean calls.
- What can be improved
- Consent was collected, but consent_id was omitted from the handoff tool.
STT, LLM and TTS
- nova-3
- openai/gpt-4.1
- sonic-3
All platform comparisons
LiveKit · ranked 2 of 8Retell · ranked 1 of 8
LiveKit vs Retell
Retell is ahead on repeatable reliability, tool call accuracy, interruption handling and response time, and LiveKit on task completion and infrastructure reliability. The two tie on voice tone and clarity.
Last updated Methodology by Luis Ojeda
Verdict
Which to pick
Pick LiveKit for completing the caller's task (1.24 points higher) and calls that connect and stay up (0.82 points higher). Pick Retell for reliability across repeated runs (4.88 points higher), accurate tool calls (0.17 higher), handling interruptions (0.03 higher) and fast replies (0.38s faster). They are level on how the agent sounds.
Head to head
Every metric, side by side
Bars share one scale across all 8 platforms. The place next to each figure is its rank in the field.
Repeatable reliability
Share of the 82 scenarios that passed on all three runs.
- LiveKit
- 70.73%2nd
- Retell
- 75.61%1st
Retell, 4.88 points higher
Task completion
Calls where the expected outcome was fully reached.
- LiveKit
- 95.12%3rd
- Retell
- 93.88%5th
LiveKit, 1.24 points higher
Infrastructure-clean calls
Calls with no connection, audio or platform failure.
- LiveKit
- 99.19%3rd
- Retell
- 98.37%4th
LiveKit, 0.82 points higher
Tool call accuracy
Mean score for calling the right tool with the right arguments.
- LiveKit
- 4.67/54th
- Retell
- 4.84/51st
Retell, 0.17 higher
Voice tone and clarity
Mean score for how clear and natural the agent sounds.
- LiveKit
- 4.36/52nd
- Retell
- 4.36/52nd
Level
Interruption handling
Mean score for yielding and recovering when the caller cuts in.
- LiveKit
- 4.97/53rd
- Retell
- 5.00/51st
Retell, 0.03 higher
Mean response time
Mean time for the agent to start replying after the caller stops.
- LiveKit
- 2.59s6th
- Retell
- 2.21s5th
Retell, 0.38s faster
246 calls per platform: 82 scenarios, 3 runs each.
On the calls
What each platform ran, and what we saw
Retell
- Strength
- Leads repeatable reliability at 75.61% pass³.
- What can be improved
- The transcript captured a phone number correctly, but a different number was sent to the tool.
STT, LLM and TTS
- accurate (Retell-managed STT)
- gpt-5.5
- eleven_flash_v2
Frequently asked questions
Is LiveKit or Retell more reliable?
Retell is more reliable, by 4.88 points. LiveKit passed 70.73% of the 82 scenarios on all 3 runs and Retell passed 75.61%. A scenario only counts when every run of it passed.
Which responds faster, LiveKit or Retell?
Retell starts replying sooner, 0.38s faster. Mean response time is 2.59s for LiveKit and 2.21s for Retell.
Which completes more calls, LiveKit or Retell?
LiveKit reached the expected outcome on 95.12% of calls and Retell on 93.88%. LiveKit is ahead on task completion.
How were LiveKit and Retell tested?
Both ran the same Appointment and Medicare agents through the same 82 caller scenarios, 3 times each, with the same evaluators and mock tools. Only the platform and its speech and model components changed.