Skip to content

LLM · Z.ai

GLM-5.3

Ranks =3 of 7 on noul, =5 of 7 on choice, =2 of 7 on score. Verbalized probabilities through the Jevals adapter, reasoning low, via OpenRouter · Novita (fp8).

compare with Jev →

noulif

60.6Decision Score
rank =3 of 8

accuracy
88.7%
calibration gap
2.0
hand-off at 95%
45%
$ per 1k
$0.51
speed p95
2.9 s
10said 55%, right 67% (3 answers)said 61%, right 71% (34 answers)said 72%, right 70% (99 answers)said 84%, right 80% (372 answers)said 94%, right 95% (992 answers)Said vs right. Squares on the diagonal mean honest confidence; bigger squares hold more answers.
choicematch

66.8Decision Score
rank =2 of 8

accuracy
78.9%
calibration gap
6.2
hand-off at 95%
38%
$ per 1k
$0.72
speed p95
3.2 s
10said 25%, right 33% (3 answers)said 33%, right 39% (38 answers)said 44%, right 43% (120 answers)said 54%, right 57% (183 answers)said 62%, right 70% (173 answers)said 73%, right 82% (283 answers)said 84%, right 91% (338 answers)said 92%, right 97% (362 answers)Said vs right. Squares on the diagonal mean honest confidence; bigger squares hold more answers.
scoresort

7.8Decision Score
rank =1 of 8

accuracy
43.0%
calibration gap
12.9
hand-off at 95%
$ per 1k
$0.78
speed p95
3.6 s
10said 35%, right 25% (4 answers)said 44%, right 24% (279 answers)said 52%, right 40% (722 answers)said 61%, right 53% (280 answers)said 71%, right 62% (130 answers)said 83%, right 67% (63 answers)said 92%, right 77% (22 answers)Said vs right. Squares on the diagonal mean honest confidence; bigger squares hold more answers.

everything behind the numbers

noul
accuracy88.7%95% range 85.391.6%
calibration gap2.0 ptslower is better · 0.3% of answers at confidence 1.00
10said 55%, right 67% (3 answers)said 61%, right 71% (34 answers)said 72%, right 70% (99 answers)said 84%, right 80% (372 answers)said 94%, right 95% (992 answers)
automation at 95%45%of decisions, acting at confidence ≥ 0.91 (right 96.7%)shared gate ≥ 0.91: acts on 45%, right 96.7%
speed1.8 smedian · p95 2.9 s
stability5.3%of answers change on an identical rerun
valid output100.0%0.0% refused · 1,500 of 1,500 answered
setupjevals-llm 0383a0e3e592 · verbalized probabilities via OpenRouter · Novita (fp8) · reasoning low (mandatory) · temperature provider default · 300 items × 5 runs · 2026-09-18 · served as z-ai/glm-5.3pubmedqa run log
choice
accuracy78.9%95% range 74.982.7%
calibration gap6.2 ptslower is better · 0.0% of answers at confidence 1.00
10said 25%, right 33% (3 answers)said 33%, right 39% (38 answers)said 44%, right 43% (120 answers)said 54%, right 57% (183 answers)said 62%, right 70% (173 answers)said 73%, right 82% (283 answers)said 84%, right 91% (338 answers)said 92%, right 97% (362 answers)
automation at 95%38%of decisions, acting at confidence ≥ 0.83 (right 95.1%)shared gate ≥ 0.96: acts on 0%, right 100.0%
speed2.1 smedian · p95 3.2 s
stability9.7%of answers change on an identical rerun · 23.7% change when options are reordered
valid output100.0%0.0% refused · 1,500 of 1,500 answered
setupjevals-llm 0383a0e3e592 · verbalized probabilities via OpenRouter · Novita (fp8) · reasoning low (mandatory) · temperature provider default · 300 items × 5 runs · 2026-09-18 · served as z-ai/glm-5.3banking77 run log
score
accuracy43.0%95% range 38.647.9%
calibration gap12.9 ptslower is better · 0.0% of answers at confidence 1.00
10said 35%, right 25% (4 answers)said 44%, right 24% (279 answers)said 52%, right 40% (722 answers)said 61%, right 53% (280 answers)said 71%, right 62% (130 answers)said 83%, right 67% (63 answers)said 92%, right 77% (22 answers)
automation at 95%never reaches 95% accuracy
speed2.2 smedian · p95 3.6 s
stability31.3%of answers change on an identical rerun
valid output100.0%0.0% refused · 1,500 of 1,500 answered
setupjevals-llm 0383a0e3e592 · verbalized probabilities via OpenRouter · Novita (fp8) · reasoning low (mandatory) · temperature provider default · 300 items × 5 runs · 2026-09-18 · served as z-ai/glm-5.3helpsteer2 run log
model id
z-ai/glm-5.3
served as
z-ai/glm-5.3
adapter
jevals-llm 0383a0e3e592 · the prompt
host
OpenRouter · Novita (fp8)
price as of
2026-09-18
harness
4da1b6d6bc57
release
2026-09-18

see every question GLM-5.3 answered →