Llama 3.3 70B by Meta scores 59.2 out of 100 on the Neura Intelligence Index, ranking #139 among all AI models. This composite score is computed from 12 individual benchmarks across coding, math, reasoning, and language categories.

Llama 3.3 70B costs $0.88/1M input tokens and $0.88/1M output tokens via API. At a typical 3:1 input-to-output ratio with 1M tokens/day usage, the estimated monthly cost is approximately $26.

Yes, Llama 3.3 70B has open weights available under the llama-3.3 license. You can download and self-host the model.

Llama 3.3 70B supports a context window of 128K tokens, allowing it to process moderate-length documents and conversations in a single request.

Llama 3.3 70B achieves its highest score on Chatbot Arena Elo with an Elo rating of 1247. This benchmark evaluates lmsys chatbot arena — crowdsourced blind comparison elo rating from human preferences.