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 2026-10-22 14:06 UTC · refreshes every few minutes

2438 Model & service calls

42.0% Succeeded

0s Median latency

2.6 Calls per graded answer grading purposes only

752048 Characters sent

10112 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	2394	98.2%	41.1%	0s	3s	633293	0
Feedback	15	0.6%	100.0%	1s	2s	32384	4817
Classification (chat model)	14	0.6%	92.9%	1s	10s	66893	561
Math checker generation	6	0.2%	33.3%	7s	45s	10760	3740
Code execution	5	0.2%	100.0%	2s	7s	3752	0
Rubric scoring	4	0.2%	100.0%	1s	1s	4966	994

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	1.1%
Python: First Steps	computer_science	10	0.4%

By endpoint

Host	Calls	Share	Succeeded	Median
speech.ai.unturf.com	2394	98.2%	41.1%	0s
qwen.ai.unturf.com	30	1.2%	100.0%	1s
hermes.ai.unturf.com	9	0.4%	44.4%	1s
unsandbox	5	0.2%	100.0%	2s

By model

tts-1-f5 2394 (98.2%)

turboderp/Qwen3.8-27B-exl3 36 (1.5%)

python 5 (0.2%)

model 3 (0.1%)

By brand

un (continuing education) 2438 (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 1083

HTTP 500 225

HTTP 401 41

HTTP 429 37

ReadTimeout 11

unavailable 6

HTTP 503 5

HTTP 502 5

ConnectionError 2