Benchmarks

Live clocks and aggregated performance metrics, updated hourly.

Prices shown are provider rates. ModelRelay adds 4.5%.

Latency
Throughput
Cost
Anthropic

Claude Haiku 4.5

Latency
0.34s
Throughput
214.8tps
Uptime
Input / 1M
$1.00
Output / 1M
$5.00
Anthropic

Claude Opus 4.5

Latency
0.79s
Throughput
74.1tps
Uptime
Input / 1M
$5.00
Output / 1M
$25.00
Anthropic

Claude Sonnet 4.5

Latency
1.79s
Throughput
89.3tps
Uptime
Input / 1M
$3.00
Output / 1M
$15.00
Google AI Studio

Gemini 3 Flash

Latency
3.77s
Throughput
202.3tps
Uptime
Input / 1M
$0.50
Output / 1M
$3.00
Google AI Studio

Gemini 3 Pro Preview

Latency
3.08s
Throughput
111.9tps
Uptime
Input / 1M
$2.00
Output / 1M
$12.00
OpenAI

GPT-5.2

Latency
0.90s
Throughput
108.9tps
Uptime
Input / 1M
$1.75
Output / 1M
$14.00
xAI

Grok 4.1 Fast

Latency
0.56s
Throughput
192.9tps
Uptime
Input / 1M
$0.20
Output / 1M
$0.50
xAI

Grok 4.1 Fast (Reasoning)

Latency
43.61s
Throughput
30.4tps
Uptime
Input / 1M
$0.20
Output / 1M
$0.50
AnthropicClaude Haiku 4.50.34s214.8tps
$1.00$5.00
AnthropicClaude Opus 4.50.79s74.1tps
$5.00$25.00
AnthropicClaude Sonnet 4.51.79s89.3tps
$3.00$15.00
Google AI StudioGemini 3 Flash3.77s202.3tps
$0.50$3.00
Google AI StudioGemini 3 Pro Preview3.08s111.9tps
$2.00$12.00
OpenAIGPT-5.20.90s108.9tps
$1.75$14.00
xAIGrok 4.1 Fast0.56s192.9tps
$0.20$0.50
xAIGrok 4.1 Fast (Reasoning)43.61s30.4tps
$0.20$0.50