BENCHMARK · MESA 0.9.2

MESA 0.9.2 benchmark report

With the built-in BGE reranker, the correct passage was first in 57.9% of answerable cases and in the top five in 74.4%. Number, date and docket questions were the weakest group at 48.3% Hit@5.

74.4%

Hit@5

Correct passage in the first five results

57.9%

Hit@1

Correct passage in the first result

0.656

MRR

Ranking of the first correct passage

MESA 0.9.2 · BGE-M3 + built-in BGE reranker (T2)

Search results

November 7, 2026 · 141 documents · 121 / 145 answerable test questions

Built-in configurations
ConfigurationHit@5Hit@1MRRHit@10Doc@1Doc@5p50p95Hardware
Models off (T0)20261107T141454Z39.7%22.3%0.30046.3%44.6%57.9%0.45 s0.94 sColab A100
Magibu embeddings (T1m)20261107T141454Z50.4%32.2%0.41059.5%52.9%68.6%1.39 s2.33 sColab A100
BGE-M3 embeddings (T1b)20261107T141454Z67.8%38.0%0.51476.9%56.2%76.9%1.35 s2.27 sColab A100

Results by question type

Hit@5 per question type. Darker cells mean the right passage was found more often.

Correct passage in the first five results
Question typeModels off (T0)Magibu embeddings (T1m)BGE-M3 embeddings (T1b)BGE-M3 + built-in BGE reranker (T2)
Exact wording (36) 75.0%77.8%86.1%97.2%
Paraphrase (39) 15.4%48.7%69.2%71.8%
Cross-document (17) 35.3%23.5%58.8%76.5%
Number, date or docket (29) 31.0%34.5%48.3%48.3%

Citation questions

Citation questions, Hit@5
ConfigurationExplicit citation (50)Natural wording (40)
Models off (T0)100.0%97.5%
Magibu embeddings (T1m)100.0%95.0%
BGE-M3 embeddings (T1b)100.0%97.5%
BGE-M3 + built-in BGE reranker (T2)100.0%100.0%

Time until searchable

Full corpus ingestion
ConfigurationTotal timeChunksHardware
Models off (T0)15.8 min5,041Colab A100
Magibu embeddings (T1m)23.7 min5,041Colab A100
BGE-M3 embeddings (T1b)20.7 min5,041Colab A100

Download results

Report and aggregate metrics (JSON) ↓

How we measured

Tier M contains 141 Turkish laws, regulations and court decisions. A language model generated questions from the corpus. A separate model and rule checks reviewed them. Configurations were chosen on 101 development questions before evaluation on 145 held-out questions. Retrieval scores use the 121 answerable questions.

What do the metrics mean?
Hit@1 · Hit@5 · Hit@10
Share of questions with the correct passage in the first 1, 5 or 10 results.
MRR
Average reciprocal rank of the first correct passage. A score of 1 means it always came first.
Doc@1 · Doc@5
Share of questions with the correct document among the first 1 or 5 results.
p50 · p95
Half of queries finished within p50 and 95% within p95. Both are measured in seconds.

Scope and limitations

The test measures whether MESA retrieves the correct passage or document. The 24 questions with no answer in the corpus are outside the retrieval scores. The corpus is small and its decisions come from the Court of Cassation. The questions are synthetic.