supermarket

Language

Llama 2 13B

By Meta, in the Llama family.

Released
18 Jul 2023
Status
Legacy
Context
4.1K
Max output
—
Input per 1M tokens
—
Output per 1M tokens
—
Open weights
Yes
License
llama-2
Parameters
13.02B
Knowledge cutoff
Sep 2022
Reasoning
No
Modalities
text → text

Category scores

Out of 100, from independent and third-party results only, by method category-2026-09-29.1.

Not enough independent results yet

Only independent and third-party results rank a model; self-reported and unattributed numbers never do.

Benchmark results

Every result we hold for this model, with who ran it and where it was published.

Knowledge & reasoning

Knowledge & reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
AI2 Reasoning Challenge (ARC)Challenge set accuracy49.4%Third party——The Falcon Series of Open Language Models via Epoch AI
AI2 Reasoning Challenge (ARC)Challenge set accuracy49.4%Third party——Gemma: Open Models Based on Gemini Research and Technology via Epoch AI
AI2 Reasoning Challenge (ARC)Challenge set accuracy48.8%Third party——Mixtral of Experts via Epoch AI
AI2 Reasoning Challenge (ARC)Challenge set accuracy60.3%Self-reported——Qwen Technical Report via Epoch AI
AI2 Reasoning Challenge (ARC)Challenge set accuracy49.4%Self-reported——Nemotron-4 15B Technical Report via Epoch AI
AI2 Reasoning Challenge (ARC)Challenge set accuracy49.4%Self-reported——Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI
BIG-Bench Hard (BBH)Average accuracy47.0%Third party——Baichuan 2: Open Large-scale Language Models via Epoch AI
BIG-Bench Hard (BBH)Average accuracy39.4%Third party——Gemma: Open Models Based on Gemini Research and Technology via Epoch AI
BIG-Bench Hard (BBH)Average accuracy39.4%Self-reported——Nemotron-4 15B Technical Report via Epoch AI
BIG-Bench Hard (BBH)Average accuracy39.4%Self-reported——Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI
HellaSwagAccuracy80.7%Third party——Gemma: Open Models Based on Gemini Research and Technology via Epoch AI
LAMBADAAccuracy76.5%Self-reported——Qwen Technical Report via Epoch AI
LAMBADAAccuracy76.5%Self-reported——Qwen Technical Report via Epoch AI
MMLUAccuracy55.6%Third party——Mixtral of Experts via Epoch AI
MMLUAccuracy55.4%Third party——Stanford CRFM Leaderboard via Epoch AI
MMLUAccuracy55.1%Third party——Baichuan 2: Open Large-scale Language Models via Epoch AI
MMLUAccuracy54.8%Third party——Gemma: Open Models Based on Gemini Research and Technology via Epoch AI
MMLUAccuracy54.8%Self-reported——Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI
MMLUAccuracy54.8%Self-reported——Nemotron-4 15B Technical Report via Epoch AI
OpenBookQAAccuracy57.0%Self-reported——Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI
PIQAAccuracy80.8%Third party——Mixtral of Experts via Epoch AI
PIQAAccuracy80.5%Third party——The Falcon Series of Open Language Models via Epoch AI
PIQAAccuracy80.5%Third party——Gemma: Open Models Based on Gemini Research and Technology via Epoch AI
PIQAAccuracy80.5%Self-reported——Qwen Technical Report via Epoch AI
PIQAAccuracy80.5%Self-reported——Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI
PIQAAccuracy79.8%Self-reported——Nemotron-4 15B Technical Report via Epoch AI
ScienceQAAccuracy55.8%Third party——ScienceQA leaderboard via Epoch AI
WinoGrandeAccuracy72.8%Third party——Gemma: Open Models Based on Gemini Research and Technology via Epoch AI
WinoGrandeAccuracy72.8%Self-reported——Nemotron-4 15B Technical Report via Epoch AI
WinoGrandeAccuracy72.8%Self-reported——Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI

Mathematics

Mathematics results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GSM8KExact match34.3%Third party——Mixtral of Experts via Epoch AI
GSM8KExact match28.9%Third party——Baichuan 2: Open Large-scale Language Models via Epoch AI
GSM8KExact match28.7%Third party——Gemma: Open Models Based on Gemini Research and Technology via Epoch AI
GSM8KExact match24.5%Third party——Stanford HELM via Epoch AI
GSM8KExact match29.6%Self-reported——Qwen Technical Report via Epoch AI
GSM8KExact match28.7%Self-reported——Nemotron-4 15B Technical Report via Epoch AI
GSM8KExact match28.7%Self-reported——Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI

Factuality & hallucination

Factuality & hallucination results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
TriviaQAExact match79.6%Third party——Gemma: Open Models Based on Gemini Research and Technology via Epoch AI
TriviaQAExact match79.6%Self-reported——Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI
TriviaQAExact match79.4%Self-reported——Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI
TriviaQAExact match77.2%Self-reported——Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI
TriviaQAExact match73.1%Self-reported——Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI

Other Llama versions

Sources