Llama 3.1 8B Instruct vs Llama 3.3 70B Instruct

Llama 3.1 8B Instruct is about 7.0× cheaper than Llama 3.3 70B Instruct on blended token cost ($0.022 vs $0.155 per 1M).

SpecLlama 3.1 8B InstructLlama 3.3 70B Instruct
ProviderMetaMeta
StatusGAGA
Input $/1M$0.02$0.1
Output $/1M$0.03$0.32
Blended $/1M$0.022$0.155
Context128K128K
Max output
Cutoff2023-122023-12