Llama 3.1 70B
Open-weight model with strong performance across many benchmarks.
Average Score
4.2Best Task
ReasoningPopularity
55%
Top Tasks
ReasoningCodingGeneration
Strengths
- Open source
- Self-hostable
- Good reasoning
Where it struggles
- Requires significant compute
- Less refined than commercial models