Codesota · Models · Llama-3.2-3B-Instructmeta-llama14 results · 3 benchmarks
Model card

Llama-3.2-3B-Instruct.

meta-llamaopen-source3.21B params
§ 01 · Benchmarks

Every benchmark Llama-3.2-3B-Instruct has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01Polish EQ-BenchNatural Language Processing · Polish Emotional Intelligenceeq-score46.2%#66/101—source ↗
02CPTU-BenchNatural Language Processing · Polish Text Understandingtricky-questions1.2%#79/93—source ↗
03CPTU-BenchNatural Language Processing · Polish Text Understandinglanguage-understanding2.3%#84/93—source ↗
04CPTU-BenchNatural Language Processing · Polish Text Understandingaverage2.0%#85/93—source ↗
05CPTU-BenchNatural Language Processing · Polish Text Understandingsentiment2.8%#85/93—source ↗
06CPTU-BenchNatural Language Processing · Polish Text Understandingphraseology1.7%#93/93—source ↗
07Open PL LLM LeaderboardNatural Language Processing · Polish LLM Generaleq-bench31.4%#138/299—source ↗
08Open PL LLM LeaderboardNatural Language Processing · Polish LLM Generaldyk46.7%#237/489—source ↗
09Open PL LLM LeaderboardNatural Language Processing · Polish LLM Generalaverage36.3%#259/491—source ↗
10Open PL LLM LeaderboardNatural Language Processing · Polish LLM Generalbelebele68.6%#292/490—source ↗
11Open PL LLM LeaderboardNatural Language Processing · Polish LLM Generalpolemo2-in69.8%#319/490—source ↗
12Open PL LLM LeaderboardNatural Language Processing · Polish LLM Generalcbd19.4%#341/490—source ↗
13Open PL LLM LeaderboardNatural Language Processing · Polish LLM Generalpolqa-open-book74.7%#365/489—source ↗
14Open PL LLM LeaderboardNatural Language Processing · Polish LLM Generalppc49.6%#405/490—source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area

Where Llama-3.2-3B-Instruct actually performs.

Natural Language Processing
3
benchmarks
avg rank #203.4
§ 04 · Related models

Other meta-llama models scored on Codesota.

Llama-3.3-70B-Instruct
70.6B params · 1 result
Llama-2-7b-chat-hf
0 results
Llama-2-7b-hf
0 results
Llama-3.2-1B
0 results
Llama-3.2-1B-Instruct
1.24B params · 0 results
Llama-3.2-3B
0 results
Llama-4-Scout-17B-16E
0 results
Llama-4-Scout-17B-16E-Instruct
109B params · 0 results
§ 05 · Sources & freshness

Where these numbers come from.

speakleash/open_pl_llm_leaderboard
8
results
SpeakLeash/CPTU-Bench
5
results
SpeakLeash/Polish-EQ-Bench
1
result
14 of 14 rows marked verified.