Model Comparison

Compare performance metrics across all tested LLM models

GPT-4-Turbo OpenAI
8.99
Quality Score
%
Conversion Rate
Avg Latency 1150 ms
P95 Latency ms
Error Rate %
Cost/1k tokens $0.01
Total Interactions 131949
Llama-3-70B Meta
8.56
Quality Score
%
Conversion Rate
Avg Latency 905 ms
P95 Latency ms
Error Rate %
Cost/1k tokens $0.0009
Total Interactions 44871
Claude-3-Opus Anthropic
9.17
Quality Score
%
Conversion Rate
Avg Latency 1346 ms
P95 Latency ms
Error Rate %
Cost/1k tokens $0.015
Total Interactions 44637
Gemini-1.5-Pro Google
8.86
Quality Score
%
Conversion Rate
Avg Latency 948 ms
P95 Latency ms
Error Rate %
Cost/1k tokens $0.007
Total Interactions 43401
GPT-3.5-Turbo OpenAI
8.29
Quality Score
%
Conversion Rate
Avg Latency 463 ms
P95 Latency ms
Error Rate %
Cost/1k tokens $0.0005
Total Interactions 43294
Mixtral-8x7B Mistral AI
8.38
Quality Score
%
Conversion Rate
Avg Latency 580 ms
P95 Latency ms
Error Rate %
Cost/1k tokens $0.0002
Total Interactions 42653
Llama-3-8B Meta
None
Quality Score
%
Conversion Rate
Avg Latency None ms
P95 Latency ms
Error Rate %
Cost/1k tokens $0.0001
Total Interactions --
Claude-3-Sonnet Anthropic
None
Quality Score
%
Conversion Rate
Avg Latency None ms
P95 Latency ms
Error Rate %
Cost/1k tokens $0.003
Total Interactions --
Quality vs Conversion
Latency Comparison
Cost-Quality Analysis
Detailed Comparison
Model Provider Quality Conversion Avg Latency P95 Latency Error Rate Cost/1k Interactions
GPT-4-Turbo OpenAI 8.99 % 1150 ms ms % $0.01 131949
Llama-3-70B Meta 8.56 % 905 ms ms % $0.0009 44871
Claude-3-Opus Anthropic 9.17 % 1346 ms ms % $0.015 44637
Gemini-1.5-Pro Google 8.86 % 948 ms ms % $0.007 43401
GPT-3.5-Turbo OpenAI 8.29 % 463 ms ms % $0.0005 43294
Mixtral-8x7B Mistral AI 8.38 % 580 ms ms % $0.0002 42653
Llama-3-8B Meta None % None ms ms % $0.0001 --
Claude-3-Sonnet Anthropic None % None ms ms % $0.003 --