Recent studyBlind TTS Elo is live. Compare two anonymous voice samples, vote after listening, and help separate real preference signal from noise.Vote in the study ->
Codesota · Models · SmoLM2 (1.7B)11 results · 11 benchmarks
Model card

SmoLM2 (1.7B).

unknown
§ 02 · Benchmarks

Every benchmark SmoLM2 (1.7B) has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01HumanEvalComputer Code · Code Generationpass-122.6%#3/3source ↗
02TriviaQANatural Language Processing · Question Answeringaccuracy36.7%#3/4source ↗
03CommonsenseQAReasoning · Commonsense Reasoningaccuracy43.6%#5/5source ↗
04Natural QuestionsNatural Language Processing · Question Answeringaccuracy8.7%#5/5source ↗
05BIG-Bench HardReasoning · Multi-step Reasoningaccuracy32.2%#11/11source ↗
06HumanEval+Computer Code · Code Generationpass-122.6%#11/12source ↗
07WinoGrandeReasoning · Commonsense Reasoningaccuracy59.4%#12/13source ↗
08HellaSwagReasoning · Commonsense Reasoningaccuracy68.7%#13/17source ↗
09MATHReasoning · Mathematical Reasoningaccuracy11.6%#46/46source ↗
10GSM8KReasoning · Mathematical Reasoningaccuracy31.1%#47/48source ↗
11MMLU-ProReasoning · Commonsense Reasoningaccuracy19.4%#71/73source ↗
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.
§ 03 · Strengths by area

Where SmoLM2 (1.7B) actually performs.

Natural Language Processing
2
benchmarks
avg rank #4.0
Computer Code
2
benchmarks
avg rank #7.0
Reasoning
7
benchmarks
avg rank #29.3
§ 04 · Papers

1 paper with results for SmoLM2 (1.7B).

  1. 2025-02-04· 11 results

    SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model

§ 06 · Sources & freshness

Where these numbers come from.

pwc-dump
11
results
0 of 11 rows marked verified.