The Model Index
Up to four models

Llama 3.3 70B vs Claude 3 Opus

On the index (GPT-4 = 100), Llama 3.3 70B scores 108 and Claude 3 Opus 108. 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 3.3 70B
Presets
Llama 3.3 70B
10892–124
WeightsOpen
ReleasedDec 2024
Price in / out$0.71 / $0.71
Context131K
Core tests taken1 of 9
Claude 3 Opus
10891–124
WeightsClosed
ReleasedMar 2024
Price in / out–
Context–
Core tests taken2 of 9

Llama 3.3 70B and Claude 3 Opus are too close to call (0 points apart, within the uncertainty), ahead on 1 of 1 shared core tests.

Test by testthird-party results

Llama 3.3 70BClaude 3 Opus
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)MMLU (HELM runs)MMLU (HELM runs)SimpleBenchSimpleBenchMATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)MATH (level 5)MATH (level 5)CODINGWeirdMLWeirdMLAGENTSDTBenchDTBenchLMCALMCA
Every result as a table
TestLlama 3.3 70BClaude 3 Opus
Reasoning & knowledge
GPQA Diamond (Epoch AI)47.4%47.2%
MMLU (HELM runs)79.1%84.6%
SimpleBench19.9%23.5%
Maths
AIME-style maths (OTIS mock)5.1%4.7%
MATH (level 5)41.6%37.5%
Coding
WeirdML14.4%19.2%
Agents
DTBench59.5%61.6%
LMCA17.5%17.0%
Task length · METR18 sec

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

This page as a chartDownload PNG