AnthropicModel

Claude Haiku 4.5

How Claude Haiku 4.5 performed across every Indonesian and regional-language benchmark we publish.Read more
Last runon Pemahaman slang Indonesiaversion not pinnedKeterangan

Benchmarks tested

4

of 6 benchmarks we publish

The rest have not been run, so this page can say nothing about how it does on 2 other benchmarks.

Runs completed

6

complete runs on record

One run means the model answered every item in a set, start to finish. A set run more than once contributes more than one.

Total cost

$0.1114

for every run above

The real amount billed by the provider, summed, not an estimate from per-token pricing. Not yet divided by correct answers.

Total tokens

50,471

input and output combined

Counted only from runs that recorded tokens. Runs on the older path did not store them, and the figure is not guessed.

How many answers were correct

Each bar is one benchmark: how many items it answered correctly out of every item in that set. We wrote the questions and answer keys ourselves and had them checked by two native speakers, and every model answers exactly the same items.

4 benchmarks
Share of items answered correctly across 4 benchmarks

How far the cost pays off

Sorted by the cheapest correct answer first. The left column shows how many items were answered correctly, so a top row with a long left bar means both cheap and accurate.

4 with cost on record
Correct answersCost per correct answer
Tingkat tutur bahasa Jawa74.1%$0.000256
Pemahaman slang Indonesia90.5%$0.000274
Arti kata Jawa sehari-hari63.7%$0.000292
Koreksi ejaan Indonesia52.6%$0.000508
Correct answers against cost per correct answer

Tokens used

Input and output are kept apart because they are priced differently, often tenfold. A long output bar means the model talks a lot, and that is where nearly all the cost difference comes from.

44,117 in, 6,354 out
inputoutput
Pemahaman slang Indonesia14,163 · 240
Tingkat tutur bahasa Jawa6,396 · 152
Arti kata Jawa sehari-hari9,041 · 5,613
Koreksi ejaan Indonesia14,517 · 349
Tokens for one full run, per benchmark

Wait time per call

Median wait for a single call on each benchmark. Median rather than mean, so one stalled call does not move the number.

4 with wait time on record
Median wait for one call, per benchmark

Test coverage

This model has been tested on 4 of the 6 benchmarks we publish. The rest have not been run, so this page cannot say anything about how it does there.

4 of 6 benchmarks we publish

Earlier question versions

The questions or the key changed after these were measured. Kept as history, not as current results.
Pemahaman slang Indonesia26/27 96.3%
Arti kata Jawa sehari-hari82/135 60.7%
Results on an earlier version of the questions or answer key
Log6 calls to this model, spread across 3 days.

This model is called through a provider id that may not pin a version, so the dates below state which weights were actually tested. Rows marked with a hyphen have a date we inferred from file history rather than recorded at call time, so we do not show a clock time for them.

1 runs
  1. 23:08Pemahaman slang Indonesiamultiple choice38/42
1 runs
  1. 21:18Arti kata Jawa sehari-hariword meaning86/135
4 runs
  1. -KeteranganArti kata Jawa sehari-hariword meaning82/135
  2. 06:42Koreksi ejaan Indonesiaopen answer20/38
  3. 06:37Pemahaman slang Indonesiamultiple choice26/27
  4. 03:59Tingkat tutur bahasa Jawamultiple choice20/27

← See all models