The Model Index
Up to four models

Claude 3.5 Haiku vs Llama 3.3 70B

On the index (GPT-4 = 100), Claude 3.5 Haiku scores 109 and Llama 3.3 70B 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 Claude 3.5 Haiku
Presets
Claude 3.5 Haiku
10993–125
WeightsClosed
ReleasedNov 2024
Price in / out–
Context–
Core tests taken1 of 9
Llama 3.3 70B
10892–124
WeightsOpen
ReleasedDec 2024
Price in / out$0.71 / $0.71
Context131K
Core tests taken1 of 9

Claude 3.5 Haiku and Llama 3.3 70B are too close to call (1 point apart, within the uncertainty), ahead on 0 of 1 shared core tests.

Test by testthird-party results

Claude 3.5 HaikuLlama 3.3 70B
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)MMLU (HELM runs)MMLU (HELM runs)MATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)MATH (level 5)MATH (level 5)CODINGWeirdMLWeirdMLAGENTSBALROGBALROGDTBenchDTBench
Every result as a table
TestClaude 3.5 HaikuLlama 3.3 70B
Reasoning & knowledge
GPQA Diamond (Epoch AI)38.1%47.4%
MMLU (HELM runs)74.3%79.1%
Maths
AIME-style maths (OTIS mock)4.3%5.1%
MATH (level 5)46.4%41.6%
Coding
WeirdML30.7%14.4%
Agents
BALROG19.3%23.0%
DTBench56.7%59.5%

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

This page as a chartDownload PNG