Cited claim

Sign in with GitHub
← Publications

Cited claim · P30 · Author-curated

On the English Open LLM Leaderboard average, Bielik-PL-Minitron-7B-v3.0-Instruct scores 67.63 and Bielik-Minitron-7B-v3.0-Instruct 66.60.

Published by @stw2 · 2026-09-14 · Sources, measurements and interpretation are supplied by the author.

Structured assertion

Relation: Metric comparison

benchmark
Open LLM LeaderboardBenchmark the values are reported on.
metric
averageScore reported on the benchmark.
language
EnglishLanguage of the evaluated tasks.
subject
Bielik-PL-Minitron-7B-v3.0-InstructModel or tokenizer after the tokenizer change.
subject value
67.63 scoreValue of the metric for the subject.
comparator
Bielik-Minitron-7B-v3.0-InstructModel or tokenizer before the tokenizer change.
comparator value
66.60 scoreValue of the metric for the comparator.

Paper citations

arXiv:2604.10799v1 →Revision supplied by author

  1. Section 5.5, p. 13

    The Polish tokenizer variants achieve 71.49 (Bielik-PL-11B-v3.0-Instruct) and 67.63 (Bielik-PL-Minitron-7B-v3.0-Instruct) on the same Open LLM Leaderboard aggregate.

  2. Table 6, p. 12, Average column, row Bielik-Minitron-7B-v3.0-Instruct (66.60)

Author’s note

Table 6 does not mark the source of the Bielik-Minitron-7B-v3.0-Instruct row; elsewhere the paper takes that model's scores from the Minitron technical report (Kinas et al. [2026]).

Concept definitions

Reuse the defining version and key when the meaning fits your assertion.

Metric comparison

The subject and the comparator take subject_value and comparator_value of the metric; benchmark, scope, evaluation_text, language and setting state what the values were measured on, where given. Values may come from separate evaluation runs.

Key metric_comparison · version 2d581473-7212-4ea2-bf80-f0c4b8cb247e

Concept JSON · Defining publication

Open LLM Leaderboard

English benchmark suite of ARC challenge, HellaSwag, WinoGrande, TruthfulQA, MMLU and GSM8K.

Key open_llm_leaderboard · version 3802da7d-eac8-4a84-b3ca-b1d8c2e12196

Concept JSON · Defining publication

Bielik-Minitron-7B-v3.0-Instruct

The 7B instruction-tuned Bielik v3 model with the original, Mistral-derived tokenizer, compressed from the 11B variant.

Key bielik_minitron_7b_v3_instruct · version 1047a8c7-2137-4bd8-b506-c9cf1390208f

Concept JSON · Defining publication

Bielik-PL-Minitron-7B-v3.0-Instruct

The 7B instruction-tuned Bielik v3 PL model with the APT4 tokenizer.

Key bielik_pl_minitron_7b_v3_instruct · version 0da1821f-a8d8-4420-aab1-d9b080ff4895

Concept JSON · Defining publication