RISE Humanities Data Benchmark, 0.6.1

Search Test Runs

 

A test run is a single execution of a benchmark test using a defined model configuration.
Each run represents how a particular large language model (LLM) — such as GPT-4, Claude-3, or Gemini — performed on a given task at a specific time, with specific settings.

A test run includes:

  • Prompt and role definition – what the model was asked to do and from what perspective (e.g. “as a historian”).
  • Model configuration – provider, model version, temperature, and other generation parameters.
  • Results – the model’s actual response and its evaluation (scores such as F1 or accuracy).
  • Usage and cost data – token counts and calculated API costs.
  • Metadata – information like the test date, benchmark name, and person who executed it.

Together, test runs make it possible to compare models, providers, and configurations across benchmarks in a transparent and reproducible way.

Loading module...

Search Results

Your search for Benchmark 'medieval_manuscripts__true' with Search Hidden 'False' returned 177 results, showing page 12 of 18.
Result 111 of 177

Test T0937 at 2026-04-21

manuscript transcription handwritten 15 de prose

openrouter qwen/qwen3.5-35b-a3b-20260224 · temp 0.0 · dataclass Document

Response vs ground truth

One model response from this run. The per-input comparison needs JavaScript.

Fuzzy score93.8%
CER ↓6.4%

{"folios":[{"folio":"3","text":"vnd ein pferit die wir vnd
minen knechten vber ruffend
sen do was meinen kün weg
deme den wir machten vnd
vnd viliend die knecht sich
vnd vil in vnz an den and
vnd die pferit vnz an die
pettel vnd was ze mol ein grofse
nebel dz wir kum gepaffend
vnd als mit groffer arbeit kome
wir ze mitttem tag zu Pant
krapffel vff den berg do
do sach ich die bücher so gar
vil herren wappen zu stone
die in stür vnd hin geben hand
do stund meines vatters pflegen
wappen och in dem zineg","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score77.4%
CER ↓24.6%

Speed

Model time, all inputs399.5 s
Mean per input39.95 s
Slowest input124.51 s
Inputs timed10

Summed model time, not elapsed: requests run in parallel and the worker count is not recorded, so the run finished sooner than this.

Costs / Pricing

Input tokens33K
Output tokens39.5K
Total tokens72.5K
Input cost$0.0054
Output cost$0.0513
Total cost$0.0567

Priced from 2026-04-21 · Input $/M $0.1625 · Output $/M $1.30

Result 112 of 177

Test T1028 at 2026-04-21

manuscript transcription handwritten 15 de prose

anthropic claude-opus-4-7 · temp 0.0 · dataclass Document

Response vs ground truth

One model response from this run. The per-input comparison needs JavaScript.

Fuzzy score97.8%
CER ↓2.1%

{"folios":[{"folio":"3","text":"vnd ein pferit die mir vnd
 minen knechten vber hulffend
 den do was mengen keyn weg
 denne den wir machtend
 vnd viehend die knecht dick
 vnd vil in vntz an den ars
 vnd die pferit vntz an die
 settel vnd was ze mol ein grosser
 nebel dz wir kum gesachend
 vnd als mit grosser arbeit kome
 wir ze mittem tag zů sant
 kristoffel vff dem berg do
 do sach ich die bůcher do gar
 vil herren wopen zu stone
 die ir stür do hyn geben hand
 do stůnd mines vatters seligen
 wopen och in dem einen","addition1":null,"addition2":null,"addition3":null}]}

Scoring

Fuzzy score50.4%
CER ↓49.6%

Speed

Model time, all inputs152.0 s
Mean per input12.67 s
Slowest input20.37 s
Inputs timed12

Summed model time, not elapsed: requests run in parallel and the worker count is not recorded, so the run finished sooner than this.

Costs / Pricing

Input tokens66.6K
Output tokens4K
Total tokens70.7K
Input cost$0.3332
Output cost$0.1008
Total cost$0.434

Priced from 2026-04-21 · Input $/M $5.00 · Output $/M $25.00

Result 113 of 177

Test T0989 at 2026-04-21

manuscript transcription handwritten 15 de prose

openrouter qwen/qwen3.5-9b-20260310 · temp 0.0 · dataclass Document

Response vs ground truth

One model response from this run. The per-input comparison needs JavaScript.

Fuzzy score93.4%
CER ↓6.6%

{"folios":[{"folio":"3","text":"Vnd ein pferit die mir vnd
minen knechtten vber helfend
sen do was menan ein weg
deme den wir machend
vnd vichend die knecht gut
vnd vil in vntz an den arb
vnd sie pferit vntz an die
pettel vnd was ze mol ein grosser
nebel do wir kum gepadend
vnd als mit grosser arbeit kame
wir ze mittem tag zu pant
empoffel vff den berg do
do sach ich die bucher do gar
vil herren wappen zu rone
die ir stur och im geber hand
do stund minne vattero pfugen
wappen set in dem emeag","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score86.4%
CER ↓15.4%

Speed

Model time, all inputs193.1 s
Mean per input48.28 s
Slowest input66.16 s
Inputs timed4

Summed model time, not elapsed: requests run in parallel and the worker count is not recorded, so the run finished sooner than this.

Costs / Pricing

Input tokens13.2K
Output tokens16.8K
Total tokens30K
Input cost$0.0013
Output cost$0.0025
Total cost$0.0038

Priced from 2026-04-21 · Input $/M $0.10 · Output $/M $0.15

Result 114 of 177

Test T0963 at 2026-04-21

manuscript transcription handwritten 15 de prose

openrouter qwen/qwen3.5-plus-20260216 · temp 0.0 · dataclass Document

Response vs ground truth

One model response from this run. The per-input comparison needs JavaScript.

Fuzzy score94.8%
CER ↓4.8%

{"folios":[{"folio":"3","text":"Vnd ein pferit die mir vnd
minen knechten vber Aulfend
Sey Do was menen tag weg
denne das wir machtend
vnd vielend die knechte dik
vnd vil in vntz an den arb
vnd die pferit vns an die
pettel vnd was ze mol ein grosse
nebel dz wir kum gepassend
vnd als mit grosser arbeit kome
wir ze mittem tag zu sant
kristoffel vff den berg do
do sach ich die bucher do gar
vil herren wapey in stone
die ir stur helm geber hand
do stund mines vatters plogen
wapey och in dem einag","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score73.4%
CER ↓28.9%

Speed

Model time, all inputs388.5 s
Mean per input32.37 s
Slowest input190.35 s
Inputs timed12

Summed model time, not elapsed: requests run in parallel and the worker count is not recorded, so the run finished sooner than this.

Costs / Pricing

Input tokens39.7K
Output tokens18.8K
Total tokens58.5K
Input cost$0.0103
Output cost$0.0293
Total cost$0.0396

Priced from 2026-04-21 · Input $/M $0.26 · Output $/M $1.56

Result 115 of 177

Test T1015 at 2026-04-21

manuscript transcription handwritten 15 de prose

openrouter google/gemma-4-31b-it-20260402 · temp 0.0 · dataclass Document

Response vs ground truth

One model response from this run. The per-input comparison needs JavaScript.

Fuzzy score90.5%
CER ↓8.4%

{"folios":[{"folio":"3","text":"Und em pferit Die wir und
minen Knechsten vber Kulsend
sey Do was menen kein Weg
denne den wir machstend
und vliehend Die Knechst Dut
und vil in vng ay sen ard
und Die pferit vng ay an Die
stat und was ze mol ein grosses
nebel Is wir kun tge passend
und als mit grosser arbeit kome
wir ze muttem tag zu sant
Kirstoffel vff den Berg do
Do fley ich Die bücher Do gar
vil herren wopen zu stone
Die in stur dclun geley hand
Do stund nuneb vatter Ruligen
wopen ach in dem emen","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score75.9%
CER ↓26.1%

Speed

Model time, all inputs297.9 s
Mean per input24.82 s
Slowest input91.23 s
Inputs timed12

Summed model time, not elapsed: requests run in parallel and the worker count is not recorded, so the run finished sooner than this.

Costs / Pricing

Input tokens9.7K
Output tokens2.9K
Total tokens12.5K
Input cost$0.0013
Output cost$0.0011
Total cost$0.0023

Priced from 2026-04-21 · Input $/M $0.13 · Output $/M $0.38

Result 116 of 177

Test T0950 at 2026-04-21

manuscript transcription handwritten 15 de prose

openrouter qwen/qwen3.5-397b-a17b-20260216 · temp 0.0 · dataclass Document

Response vs ground truth

One model response from this run. The per-input comparison needs JavaScript.

Fuzzy score94.8%
CER ↓5.0%

{"folios":[{"folio":"3","text":"Vnd ein pferit die mir vnd
minen knechten vber Rulfend
sey do was menig böser weg
deme den wir machtend
vnd viehend die knehte dick
vnd vil in vntz an den arb
vnd die pferit vntz an die
pettel vnd was ze mol ein grosse
nebel dz wir kum gepassend
vnd also mit grosser arbeit kame
wir ze mittem tag zu sant
kristoffel vff den berg do
do sach ich die bücher do gar
vil herren wapey zu ross
die ir stür helm gebey hand
do stünd mines vatters seligen
wapey och in dem einen","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score72.9%
CER ↓28.3%

Speed

Model time, all inputs989.7 s
Mean per input82.47 s
Slowest input299.59 s
Inputs timed12

Summed model time, not elapsed: requests run in parallel and the worker count is not recorded, so the run finished sooner than this.

Costs / Pricing

Input tokens39.6K
Output tokens44K
Total tokens83.6K
Input cost$0.0154
Output cost$0.103
Total cost$0.1184

Priced from 2026-04-21 · Input $/M $0.39 · Output $/M $2.34

Result 117 of 177

Test T0911 at 2026-04-21

manuscript transcription handwritten 15 de prose

openrouter qwen/qwen3.5-122b-a10b-20260224 · temp 0.0 · dataclass Document

Response vs ground truth

One model response from this run. The per-input comparison needs JavaScript.

Fuzzy score93.8%
CER ↓6.1%

{"folios":[{"folio":"3","text":"Und ein pferit die mir und
minen knechten wider gulfend
sey so was menen kein weg
deme den wir machtend
und vieliend die knecht gut
und vil in vntz an den arb
und die pferit vntz an die
pettel und was ze mol ein grosse
nebel dz wir kum gepatend
und als mit grosser arbait kome
wir ze mittern tag zu Sant
kristoffel vff den berg do
do sach ich die bücher do gar
vil herrey wopen zu ston
die ir fürer ochin gebey hand
do pünd minnes vatters pügen
wopen och in dem einen","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score77.2%
CER ↓24.6%

Speed

Model time, all inputs72.8 s
Mean per input6.07 s
Slowest input10.89 s
Inputs timed12

Summed model time, not elapsed: requests run in parallel and the worker count is not recorded, so the run finished sooner than this.

Costs / Pricing

Input tokens39.7K
Output tokens2.8K
Total tokens42.5K
Input cost$0.0103
Output cost$0.0058
Total cost$0.0162

Priced from 2026-04-21 · Input $/M $0.26 · Output $/M $2.08

Result 118 of 177

Test T1002 at 2026-04-21

manuscript transcription handwritten 15 de prose

openrouter google/gemma-4-26b-a4b-it-20260403 · temp 0.0 · dataclass Document

Response vs ground truth

One model response from this run. The per-input comparison needs JavaScript.

Fuzzy score91.9%
CER ↓8.2%

{"folios":[{"folio":"3","text":"und ein pferit die wir vnd
mine knestet wider kulfend
sen do was menen kün weg
jenne den wir machtend
und viehend die knestet dük
und vil in vng an den and
und die pferit vng an die
stuel vnd was ze mol ein grosser
nebel dz wir kun gegepende
und als mit grosser arbeit kome
wir ze nuttem tag zu sant
kerstoffel off den bery do
do hach ich die bücher do gar
vil herren wopen in stene
die in stur delym geley hand
do stünd nunne vatter flugen
wopen acht in dem enem","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score71.5%
CER ↓32.6%

Speed

Model time, all inputs80.7 s
Mean per input6.72 s
Slowest input10.59 s
Inputs timed12

Summed model time, not elapsed: requests run in parallel and the worker count is not recorded, so the run finished sooner than this.

Costs / Pricing

Input tokens9.7K
Output tokens2.7K
Total tokens12.4K
Input cost$0.0008
Output cost$0.0009
Total cost$0.0017

Priced from 2026-04-21 · Input $/M $0.08 · Output $/M $0.35

Result 119 of 177

Test T0976 at 2026-04-21

manuscript transcription handwritten 15 de prose

openrouter qwen/qwen3.5-flash-20260224 · temp 0.0 · dataclass Document

Response vs ground truth

One model response from this run. The per-input comparison needs JavaScript.

Fuzzy score94.1%
CER ↓6.1%

{"folios":[{"folio":"3","text":"vnd ein pferit die wir vnd
minen knechten vber hulfend
sen do was meinen kün weg
deme den wir machten vnd
vnd viliend die knecht sich
vnd vil in vnz an den and
vnd die pferit vnz an die
pettel vnd was ze mol ein grofse
nebel dz wir kum gepaffend
vnd als mit groffer arbeit kome
wir ze mittem tag zu pant
krapffel vff den berg do
do sach ich die bücher so gar
vil herren wopen zu stone
die in stür vnd gun geben hand
do stund mines vatter vß flugen
wopen och in dem zineg","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score76.3%
CER ↓25.4%

Speed

Model time, all inputs46.2 s
Mean per input4.20 s
Slowest input4.92 s
Inputs timed11

Summed model time, not elapsed: requests run in parallel and the worker count is not recorded, so the run finished sooner than this.

Costs / Pricing

Input tokens36.4K
Output tokens2.5K
Total tokens38.9K
Input cost$0.0024
Output cost$0.0006
Total cost$0.003

Priced from 2026-04-21 · Input $/M $0.065 · Output $/M $0.26

Result 120 of 177

Test T0898 at 2026-04-21

manuscript transcription handwritten 15 de prose

openrouter qwen/qwen3.6-plus-04-02 · temp 0.0 · dataclass Document

Response vs ground truth

One model response from this run. The per-input comparison needs JavaScript.

Fuzzy score61.8%
CER ↓38.8%

{"folios":[{"folio":"73","text":"It zu dem ersten min her graff albrecht vo
nberg
It her ludwig lantgraf von liechtenberg
It graff ott von hemmenberg
It graff Bernhart von Regenstein
It graff gottfrid von hohenloch
It graff gebhart von Mansfeld
It graff ludwig von Arenburg
It her Gottfrid von Stoff
It her hans von tungern
It Jerg Bappanhein
It her Ruprecht vom stein wz vorhin Ritter
It hug vom Rotenhan
It Jerg von waldenfels
It hilprand von thungen
It Jerg von Elicben
It henyn quast","addition1":"bedpanckin","addition2":"","addition3":""}]}

Scoring

Fuzzy score65.3%
CER ↓35.5%

Speed

Model time, all inputs655.6 s
Mean per input72.84 s
Slowest input122.63 s
Inputs timed9

Summed model time, not elapsed: requests run in parallel and the worker count is not recorded, so the run finished sooner than this.

Costs / Pricing

Input tokens29.7K
Output tokens49.8K
Total tokens79.6K
Input cost$0.0097
Output cost$0.0972
Total cost$0.1069

Priced from 2026-04-21 · Input $/M $0.325 · Output $/M $1.95