AIPing.tech | Dashboard

Live Overview Cards

Compare

Gateway Tax = (gateway - direct).

Model A

Model B

Auto-pick: Direct + GW Mini trend · last samples

How We Measure

Methodology

Each latency figure comes from full chat prompts sent to the provider; we wait for streamed completions, capturing provider overhead, the model's thinking time, and transport.

Samples aggregate real conversation flows, not pings.
Gateway tax = gateway latency - direct latency.
Freshness badges combine sample count and last-run timestamps.

Rankings

Avg heute (UTC)

Top 10 · min 3 samples

#1 Groq Direct meta-llama/llama-4-scout-17b-16e-instruct

n=24

97 ms

#2 Groq Direct openai/gpt-oss-120b

n=23

117 ms

#3 Groq Direct openai/gpt-oss-20b

n=36

130 ms

#4 Mistral Direct open-mistral-7b

n=23

169 ms

#5 Groq Direct llama-3.1-8b-instant

n=24

179 ms

#6 Mistral Direct voxtral-mini-latest

n=24

181 ms

#7 Groq Direct openai/gpt-oss-safeguard-20b

n=23

189 ms

#8 Mistral Direct voxtral-small-latest

n=35

190 ms

#9 Groq Direct groq/compound

n=24

194 ms

#10 Mistral Direct voxtral-small-2507

n=35

201 ms

Schnellste Messung (letzte 20m)

Top 10 · min 1 sample

#1 Groq Direct meta-llama/llama-4-scout-17b-16e-instruct

n=7

64 ms

#2 Groq Direct openai/gpt-oss-20b

n=10

67 ms

#3 Groq Direct openai/gpt-oss-120b

n=6

93 ms

#4 Groq Direct groq/compound

n=7

125 ms

#5 Groq Direct llama-3.1-8b-instant

n=7

134 ms

#6 Mistral Direct voxtral-mini-latest

n=7

135 ms

#7 Mistral Direct open-mistral-7b

n=6

135 ms

#8 Groq Direct meta-llama/llama-4-maverick-17b-128e-instruct

n=10

137 ms

#9 Mistral Direct voxtral-small-2507

n=9

137 ms

#10 Groq Direct openai/gpt-oss-safeguard-20b

n=6

138 ms

Open full Rankings →

Latency Trend

Legend

858 ms

AVG LATENCY

132 ms

MIN LATENCY

13958 ms

MAX LATENCY

1299

SAMPLES

API Server Map

Live endpoints

Top regions

Data lädt …

Markers show observed API endpoint IPs (GeoIP cached).

We measure from Germany; our server is pinned on the map.