Analytics

Public & anonymized. Same page for everyone, no account needed. Rows hold shape, never identity: no ip, no cookie id, no user id, no answer text. Cells under 7 observations stay hidden. How this data works · JSON

un (continuing education) · Last 28 days · computed 2026-10-19 08:53 UTC · refreshes every few minutes

2807 Model & service calls

29.8% Succeeded

0s Median latency

2.64 Calls per graded answer grading purposes only

947279 Characters sent

9816 Characters received

Every outbound call to a language model, the decision classifier, translation, speech, lesson generation or code execution. Purpose, host, latency & outcome only. No prompt or completion text.

By purpose

Purpose	Calls	Share	Succeeded	Median	95th	Chars in	Chars out
Text to speech	2766	98.5%	29.0%	0s	3s	833682	0
Feedback	14	0.5%	100.0%	1s	2s	30926	4561
Classification (chat model)	13	0.5%	92.3%	0s	10s	63204	521
Math checker generation	6	0.2%	33.3%	7s	45s	10760	3740
Rubric scoring	4	0.1%	100.0%	1s	1s	4966	994
Code execution	4	0.1%	100.0%	2s	7s	3741	0

By lesson

Which lessons drove model calls — chat, feedback and speech attributed to the lesson the learner was on.

Lesson	Subject	Calls	Share
Geometry of Site Reliability Engineering	computer_science	26	0.9%
Python: First Steps	computer_science	7	0.2%

By endpoint

Host	Calls	Share	Succeeded	Median
speech.ai.unturf.com	2766	98.5%	29.0%	0s
qwen.ai.unturf.com	29	1.0%	100.0%	1s
hermes.ai.unturf.com	8	0.3%	37.5%	1s
unsandbox	4	0.1%	100.0%	2s

By model

tts-1-f5 2766 (98.5%)

turboderp/Qwen3.8-27B-exl3 34 (1.2%)

python 4 (0.1%)

model 3 (0.1%)

By brand

un (continuing education) 2807 (100.0%)

Calls over time

Failed calls over time

Calls over time by purpose

Calls: classify

Calls: feedback

Calls: rubric

Calls: tts

Failure reasons

HTTPError 1693

HTTP 500 225

HTTP 429 29

ReadTimeout 11

HTTP 502 5

unavailable 4

ConnectionError 2

HTTP 503 1