The Model Index
Up to four models

Llama 3.2 vs GPT-4o mini

On the index (GPT-4 = 100), Llama 3.2 scores 106 and GPT-4o mini 104. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to Llama 3.2
Presets
Llama 3.2
10690–122
WeightsOpen
ReleasedSep 2024
Price in / out–
Context–
Core tests taken1 of 9
GPT-4o mini
10487–120
WeightsClosed
ReleasedJul 2024
Price in / out$0.15 / $0.60
Context128K
Core tests taken3 of 9

Llama 3.2 and GPT-4o mini are too close to call (2 points apart, within the uncertainty), ahead on 1 of 1 shared core tests.

Test by testthird-party results

Llama 3.2GPT-4o mini
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)MMLU (HELM runs)MMLU (HELM runs)GeoBenchGeoBenchMATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)MATH (level 5)MATH (level 5)AGENTSBALROGBALROG
Every result as a table
TestLlama 3.2GPT-4o mini
Reasoning & knowledge
GPQA Diamond (Epoch AI)41.0%37.7%
MMLU (HELM runs)80.3%76.7%
GeoBench52.0%64.0%
Maths
AIME-style maths (OTIS mock)2.6%6.9%
MATH (level 5)39.4%52.6%
Agents
BALROG27.3%17.4%

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG