Comprehensive side-by-side LLM comparison
Llama 3.2 90B Instruct leads with 9.4% higher average benchmark score. Llama 3.2 11B Instruct is $0.65 cheaper per million tokens. Overall, Llama 3.2 90B Instruct is the stronger choice for coding tasks.
Meta
Llama 3.2 11B was introduced as a mid-sized variant in the Llama 3.2 family, designed to offer enhanced capabilities while maintaining efficiency. Built to provide a balanced option for applications requiring more than lightweight models but less than flagship sizes, it serves diverse use cases in the open-source community.
Meta
Llama 3.2 90B was developed as a flagship-tier open-source model, designed to provide advanced capabilities with 90 billion parameters. Built to serve applications requiring high-quality reasoning and generation, it represents a powerful option within the Llama 3.2 series for demanding tasks.
Launched on the same date

Llama 3.2 11B Instruct
Meta
2024-09-25

Llama 3.2 90B Instruct
Meta
2024-09-25
Cost per million tokens (USD)

Llama 3.2 11B Instruct

Llama 3.2 90B Instruct
Context window and performance specifications
Average performance across 10 common benchmarks

Llama 3.2 11B Instruct

Llama 3.2 90B Instruct
Llama 3.2 11B Instruct
2023-12-31
Available providers and their performance metrics

Llama 3.2 11B Instruct
Bedrock
DeepInfra
Fireworks
Groq
Sambanova

Llama 3.2 11B Instruct

Llama 3.2 90B Instruct

Llama 3.2 11B Instruct

Llama 3.2 90B Instruct
Together

Llama 3.2 90B Instruct
Bedrock
DeepInfra
Fireworks
Hyperbolic
Together