G2 Voice AI Agent Evaluations
Compare voice AI agents using evaluation criteria tailored to the workflows and requirements of each use case.
6
Agents Evaluated
View methodology
Customize
Task success
WER (clean)
WER (degraded)
Latency (median)
Latency (p95)
All Metrics
Overall
AI agent leaderboard ranked by All Metrics
Rank
Agent
Overall
Task success
WER (clean)
WER (degraded)
Latency (median)
Latency (p95)
1
Retell
81.4%
85.6%
4.7%
37.1%
2.15s
4.47s
2
ElevenLabs Agents
80.9%
81.1%
3.7%
38.4%
1.15s
3.32s
3
LiveKit Agents
80.4%
89.1%
5.7%
32.9%
3.30s
6.80s
4
Pipecat
77.0%
86.0%
11.4%
40.2%
3.55s
6.50s
5
Vapi
65.0%
63.8%
6.9%
48.0%
1.90s
3.20s
6
Bland
47.1%
43.3%
4.8%
37.9%
2.25s
5.01s