Benchmark Almanac

Barcelona Supercomputing Center

We track 17 benchmarks that Barcelona Supercomputing Center made or helped make.

0 in use 11 with public code 2 with a leaderboard here

Benchmarks

Name Kind Released Best score Status
NotSoTiny 25-12 Models reconstruct a missing Verilog module from surrounding hardware design code. A formal checker compares the result with the original module. Electronics Dec 2025 31.6% Not in cards
RuC-tt07-32k Models fill missing Verilog statements in Tiny Tapeout designs. A parser checks syntax, and a formal checker tests whether the completed design preserves behavior. Electronics Apr 2026 76.7% Not in cards
MultiClinSum Models summarize clinical case reports in English, Spanish, French, and Portuguese. Organizers compare each summary with a reference using ROUGE and BERTScore. Medicine Apr 2025 0.86 Not in cards
caBreu Models summarize Catalan news articles. Reference summaries cover extractive, abstractive, and very short summaries. Reading in many languages Sep 2023 Not in cards
CatalanQA Models extract answers from Catalan passages drawn from VilaQuAD and ViquiQuAD. The dataset balances question types. Reading in many languages Jun 2022 Not in cards
CatCoLA Models classify Catalan sentences by whether a native speaker would consider them grammatical. Reading in many languages May 2024 Not in cards
CoQCat Models answer follow-up questions in Catalan using a passage and the preceding conversation. Reading in many languages Sep 2023 Not in cards
DisTEMIST Models locate disease mentions in Spanish clinical cases and link them to SNOMED CT codes. The organizers score both tasks with microaveraged F1. Medicine Dec 2021 Not in cards
EsCoLA Models judge the grammatical acceptability of Spanish sentences taken from linguistic publications. Reading in many languages May 2024 Not in cards
MEDDOPROF Models find occupations and employment status in Spanish clinical notes. They identify whose occupation each mention describes and assign an occupational or clinical code. Medicine Dec 2020 Not in cards
MedProcNER Models identify procedures in Spanish clinical cases, link them to SNOMED CT codes, and list the codes for each case. Medicine Jan 2023 Not in cards
MultiCardioNER Models mark disease mentions in Spanish cardiology cases and medication mentions in Spanish, English, and Italian versions of the cases. Organizers score exact spans with microaveraged F1. Medicine Apr 2024 Not in cards
MultiClinAI Models identify diseases, symptoms, and procedures in seven languages. A second task transfers annotated Spanish mentions to the corresponding spans in six translations. Medicine Nov 2025 Not in cards
MultiClinSum-2 Models summarize 1,000 clinical cases per language across 15 languages. Organizers report four automatic metrics and four language model judgments separately. Medicine Feb 2026 Not in cards
Parafraseja Models decide whether two Catalan sentences express the same meaning. Human annotators wrote contrasting pairs. Reading in many languages Oct 2022 Not in cards
SympTEMIST Models locate symptoms, signs, and findings in Spanish clinical cases. They also map given mentions to SNOMED CT codes in Spanish and five additional languages. Medicine Jul 2023 Not in cards
TE-ca Models decide whether a Catalan hypothesis follows from a premise. The reference labels identify the inference relation. Reading in many languages Jul 2021 Not in cards