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 10 of 18.
Result 91 of 177

Test T1411 at 2026-07-25

manuscript transcription handwritten 15 de prose

anthropic claude-opus-5 · temp 0.0 · dataclass Document

Response vs ground truth

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

Fuzzy score97.9%
CER ↓1.1%

{"folios":[{"folio":"3","text":"vnd ein pferit die mir vnd
minen knechten vber hulffend
den do was menden kein 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 zuo sant
cristoffel vff den berg do
do sach ich die buecher do gar
vil herren wopen in stond
die ir stür dohin geben hand
do stund mines vatters seligen
wopen och in dem einen","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score64.2%
CER ↓35.9%

Speed

Model time, all inputs261.0 s
Mean per input21.75 s
Slowest input39.85 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 tokens61.8K
Output tokens4K
Total tokens65.8K
Input cost$0.3091
Output cost$0.1007
Total cost$0.4098

Priced from 2026-07-25 · Input $/M $5.00 · Output $/M $25.00

Result 92 of 177

Test T1351 at 2026-07-25

manuscript transcription handwritten 15 de prose

openai gpt-5.6-sol · 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 ↓4.6%

{"folios":[{"folio":"3","text":"vnd ein pfferit Die mir vnd
minen knechten wider hulffend
Daz Do was menchen lang weg
danne Daz wir nuechtend
vnd vielend die knechte dick
vnd vil in vns an den arß
vnd die pfferit was an die
kettel vnd was ze mol ein grosser
nebel dz wir kaum gesachend
vnd also mit grosser arbeit kamē
wir ze mittem tag zu sant
kristoffel vff den berg Do
Do sach ich die buecher Do gar
vil herren woppen zu stond
Die ir stuer Do hin geben hand
Do stund mines vatters seligen
woppen och in dem einen","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score81.4%
CER ↓19.3%

Speed

Model time, all inputs812.9 s
Mean per input67.74 s
Slowest input95.50 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 tokens44.4K
Output tokens43.3K
Total tokens87.8K
Input cost$0.2222
Output cost$1.30
Total cost$1.52

Priced from 2026-07-25 · Input $/M $5.00 · Output $/M $30.00

Result 93 of 177

Test T1366 at 2026-07-25

manuscript transcription handwritten 15 de prose

openai gpt-5.6-terra · temp 0.0 · dataclass Document

Response vs ground truth

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

Fuzzy score91.6%
CER ↓9.2%

{"folios":[{"folio":"3","text":"Vnd ein pferit die mir vnd
minen knechten vber hulffend
Jm so was menchen bergweg
zeme den wir machten
vnd inckends die knechte dick
vnd vil in vns angieng ans
vnd die pferit vns in die
pittel vnd vns ze mol ein grosse
nebel dÿ wir kum gespachend
vnd als mit grossen arbeit kamen
wir ze mitten tag zu sant-
Kunstaffel vff den berg do
Do hieß ich die hütten Do gar
mit herren wopen zu hant
die in für schlin gelüein hand
do fünff mines vatters slagen
wopen ach in dem einen","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score69.6%
CER ↓35.3%

Speed

Model time, all inputs340.5 s
Mean per input28.37 s
Slowest input40.11 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 tokens44.4K
Output tokens25.7K
Total tokens70.2K
Input cost$0.1111
Output cost$0.3859
Total cost$0.497

Priced from 2026-07-25 · Input $/M $2.50 · Output $/M $15.00

Result 94 of 177

Test T1426 at 2026-07-25

manuscript transcription handwritten 15 de prose

genai gemini-3.6-flash · temp 0.0 · dataclass Document

Response vs ground truth

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

Fuzzy score96.4%
CER ↓3.3%

{"folios":[{"folio":"3","text":"Vnd ein pferit die mir vnd
minen knechten über hulfend
den do was menan kein weg
denne den wir machtend
vnd vichend die knecht dur
vnd vil in vntz an den arb
vnd die pferit vntz an die
settel vnd was ze mol ein grosser
nebel dz wir küm gesehend
vnd als mit grosser arbeit kamē
wir ze mittem tag zü sant
kristoffel uff den berg do
do sach ich die bücher so gar
vil herren wappen in stend
die ir stür dahin geben hand
do stund mines vatters seligen
wappen och in dem einen","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score78.3%
CER ↓22.3%

Speed

Model time, all inputs510.1 s
Mean per input42.51 s
Slowest input77.13 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 tokens19.1K
Output tokens2.8K
Total tokens21.9K
Input cost$0.0287
Output cost$0.0209
Total cost$0.2358
Reasoning cost usd$0.1862
Total reasoning tokens24.8K

Priced from 2026-07-25 · Input $/M $1.50 · Output $/M $7.50

Result 95 of 177

Test T1196 at 2026-07-02

manuscript transcription handwritten 15 de prose

anthropic claude-fable-5 · temp 0.0 · dataclass Document

Response vs ground truth

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

Fuzzy score98.0%
CER ↓0.9%

{"folios":[{"folio":"3","text":"vnd ein pferit die mir vnd
minen knechten vber hulfend
den do was menan kein weg
denne den wir machtend
vnd vielend die knecht dik
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 also mit grosser arbeit kame
wir ze mittem tag zuo sant
kristoffel vff den berg do
do sach ich die buecher do gar
vil herren wopen in stone
die ir stur de hin geben hand
do stuond mines vatters seligen
wopen och in dem einen","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score87.1%
CER ↓12.7%

Speed

Model time, all inputs147.5 s
Mean per input12.29 s
Slowest input18.49 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 tokens61.9K
Output tokens4.2K
Total tokens66K
Input cost$0.6186
Output cost$0.2078
Total cost$0.8264

Priced from 2026-07-02 · Input $/M $10.00 · Output $/M $50.00

Result 96 of 177

Test T1183 at 2026-07-01

manuscript transcription handwritten 15 de prose

anthropic claude-sonnet-5 · temp 0.0 · dataclass Document

Response vs ground truth

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

Fuzzy score0.0%
CER ↓100.0%

{"query": {"folios": [{"folio": "3", "text": "vnd ein pferit die mir vnd
minen knechten wor gulteng
was do was menck kein weg
denne den wir machtend
vnd viehend die knecht dick
vnd vil in vntz an den ar\u00df
vnd die pferit vntz an die
settel vnd was ze mol ein grosser
nebel dz wir kum ge sahend
vnd als wir grosser arbeit kome
wir ze mittem tago zu sant
kristoffel vff den berg do
Do sach ich die bucher do gar
vil herren wopen ju stone
die ir st\u00fcr de key gebey hand
Eo stund mines vatters selgen
wopen och in dem einen", "addition1": "", "addition2": "", "addition3": ""}]}}

Scoring

Fuzzy score0.0%
CER ↓100.0%

Speed

Model time, all inputs114.0 s
Mean per input9.50 s
Slowest input18.30 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 tokens62.6K
Output tokens4.2K
Total tokens66.8K
Input cost$0.1253
Output cost$0.0416
Total cost$0.1669

Priced from 2026-07-01 · Input $/M $2.00 · Output $/M $10.00

Result 97 of 177

Test T1170 at 2026-06-29

manuscript transcription handwritten 15 de prose

genai gemini-3.1-flash-lite · temp 0.0 · dataclass Document

Response vs ground truth

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

Fuzzy score95.3%
CER ↓4.1%

{"folios":[{"folio":"3","text":"Vnd ein pferit die mir vnd
minen knechten ibex hulfend
sen do was menen kein weg
denne den wir machtend
und viehend die knecht dut
vnd vil in vns an den arb
vnd die pferit vns an die
petel vnd was ze mol ein grosse
nebel dz wir kum gepassend
vnd als mit grosser arbeit kame
wir ze mittem tag zu sant
Kristoffel vff den berg do
do sach ich die bucher do gar
vil herren wopen in stont
die in stur delhu geben hand
do stund mines vatters rigen
wopen ach in dem einen","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score74.5%
CER ↓26.3%

Speed

Model time, all inputs32.3 s
Mean per input2.69 s
Slowest input3.77 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 tokens19.1K
Output tokens2.7K
Total tokens21.9K
Input cost$0.0048
Output cost$0.0041
Total cost$0.0089

Priced from 2026-06-29 · Input $/M $0.25 · Output $/M $1.50

Result 98 of 177

Test T1108 at 2026-06-08

manuscript transcription handwritten 15 de prose

x-ai grok-4.3 · temp 0.0 · dataclass Document

Response vs ground truth

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

Fuzzy score94.9%
CER ↓5.3%

{"folios":[{"folio":"3","text":"Und ein pferit die wir vnd
minen knechten vber Růfend
den So was menan kein Weg
denne den wir machtend
vnd vriechend die knechte vnd
vnd vil in vntz an den anb
vnd die pferit vntz an die
petrel vnd was ze mol ein grosser
nebel dz wir kum gepaſsend
vnd als mit grosser arbeit kome
wir ze mittem tag zů sant
kůnratel vff den berg So
So sach ich die bucher So gar
vil herren wapen jn ston
die ir stůr do hin geben hand
So stůnd mines vatters Růgen
wapen ouch in dem einen","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score66.7%
CER ↓34.4%

Speed

Model time, all inputs247.8 s
Mean per input20.65 s
Slowest input39.55 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 tokens38.4K
Output tokens2.6K
Total tokens41K
Input cost$0.048
Output cost$0.0065
Total cost$0.0814
Reasoning cost usd$0.0269
Total reasoning tokens10.8K

Priced from 2026-06-05 · Input $/M $1.25 · Output $/M $2.50

Result 99 of 177

Test T1121 at 2026-06-05

manuscript transcription handwritten 15 de prose

openrouter meta-llama/llama-4-scout-17b-16e-instruct · temp 0.0 · dataclass Document

Response vs ground truth

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

Fuzzy score89.3%
CER ↓12.6%

{"folios":[{"folio":"3","text":"Vnd ein pferit die wir vnd
minen freynden wolke auffend
fen So was menniglich ein weg
Semme den wir machten
vnd vilkend die Entge snd
vnd vil in vnitz an dem amb
vnd die pferit vntz an die
tertel vnd war ze mol von strotex
nebel das wir kum gesporkend
vnd als mit groffer arbeit kome
wir ze mittelm tag zu fant
Ein koffel vff dem Berg So
fach ich die bicher So far
vil horen in wapen von deme
die wir von Eckm geliky band
so tunt mines vatters Rligen
wopen sag in dem sinen","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score56.8%
CER ↓47.8%

Speed

Model time, all inputs845.5 s
Mean per input70.46 s
Slowest input93.57 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 tokens33.9K
Output tokens2.7K
Total tokens36.5K
Input cost$0.0027
Output cost$0.0008
Total cost$0.0035

Priced from 2026-06-05 · Input $/M $0.08 · Output $/M $0.30

Result 100 of 177

Test T1134 at 2026-06-05

manuscript transcription handwritten 15 de prose

openrouter stepfun/step-3.7-flash-20260528 · temp 0.0 · dataclass Document

Response vs ground truth

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

Fuzzy score90.3%
CER ↓10.3%

{"folios":[{"folio":"3","text":"Und ein pferit die mir und
minen enckten wider zulffend
den so was minen son weg
same den wir machtend
und sihend die enckte dut
und vil in vntgansen arb
und die pferit vntg an die
puttel und was ze mal ein groffer
nebel dz win kum gespottend
und als mit groffer arbeit kome
win ze mittem tag zu furt
kinstossel vff das berg so
so fach ich die bucher so gar
vil herren wopen in bone
die in fur sdin gebin gand
so fund minen vatter ob ligen
wopen und in sem einen","addition1":"","addition2":"","addition3":""}]}

Scoring

Fuzzy score64.3%
CER ↓42.0%

Speed

Model time, all inputs1786.8 s
Mean per input148.90 s
Slowest input233.61 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 tokens20.2K
Output tokens114.1K
Total tokens134.3K
Input cost$0.004
Output cost$0.1312
Total cost$0.1352

Priced from 2026-06-05 · Input $/M $0.20 · Output $/M $1.15