supermarket

Language

Llama 3.2 90B Vision

By Meta, in the Llama family.

Released
25 Sep 2024
Status
Legacy
Context
131.1K
Max output
—
Input per 1M tokens
—
Output per 1M tokens
—
Open weights
Yes
License
llama-3.2
Parameters
88.59B
Knowledge cutoff
Dec 2023
Reasoning
No
Modalities
text, image → text

Category scores

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

Category scores
CategoryScoreRankCoverage
Agentic tool use27.3Not ranked1 of 2

Benchmark results

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

Knowledge & reasoning

Knowledge & reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
MMLUAccuracy80.3%Third party——Stanford CRFM Leaderboard via Epoch AI

Scientific reasoning

Scientific reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GPQA DiamondAccuracy41.0%Independent—27 Jan 2025Epoch AI
GPQA DiamondAccuracy46.1%Self-reported——EvalEval via Hugging Face Community Evals

Mathematics

Mathematics results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
MATH Level 5Accuracy39.4%Independent—27 Jan 2025Epoch AI
OTIS Mock AIME 2024-2025Accuracy2.6%Independent—25 Feb 2025Epoch AI

Agentic tool use

Agentic tool use results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
BALROGAverage progress27.3%Third party——https://balrogai.com/ via Epoch AI

Vision & multimodal understanding

Vision & multimodal understanding results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GeoBenchCountry accuracy52.0%Third party——GeoBench leaderboard via Epoch AI

Other Llama versions

Sources