Skip to main content
Meta·current

Llama 3.3 70B

Meta's efficient open model — 405B performance in a 70B package

ProviderMeta
Released2024-12-06
StatusCurrent
Context window131K tokens
PricingOpen
Capabilities
textcodefunction-calling
Hugging FaceView on HF ↗

Benchmarks

MMLU (knowledge)86.0
MATH (mathematics)77.0

As of 2025-01-01. Source

Loading…

Compare with similar models

Llama 3.3 70BDeepSeek V3Llama 3.1 405BDeepSeek R1
Context131K66K131K66K
MMLU86.088.588.690.8
HumanEval91.6
MATH77.090.273.597.3
GPQA71.5
PricingOpenOpenOpenOpen
Input $/M$0.27$0.55

Compare more models →