Home/Models/Llama 3.1 405B

Llama 3.1 405B

Open Source
by Meta· Released 2024-07
Official Docs ↗

Llama 3.1 405B is Meta's largest open-source model, competitive with GPT-4o on many tasks. As a fully open-weight model, it can be self-hosted for complete data control, making it popular in enterprise environments with strict data privacy requirements.

Input Price
$2.70/1M
per 1M tokens
Output Price
$2.70/1M
per 1M tokens
Context Window
131K tokens
max tokens
ELO Score
1267
LMSYS Arena

Pricing Breakdown

VolumeInput CostOutput CostCombined (50/50)
1,000 tokens$0.0027$0.0027$0.0027
10,000 tokens$0.0270$0.0270$0.0270
100,000 tokens$0.2700$0.2700$0.2700
1,000,000 tokens$2.7000$2.7000$2.7000

Strengths

  • Open source — can be self-hosted for data privacy
  • Competitive with GPT-4o on many benchmarks
  • Strong multilingual capabilities
  • No output tokens premium — flat pricing

Weaknesses

  • Expensive to self-host at 405B scale
  • Slightly behind frontier closed models on reasoning

Best For

codingresearchtranslationreasoning

Compare Llama 3.1 405B With

Llama 3.1 405B vs o3
OpenAI
Llama 3.1 405B vs DeepSeek R1
DeepSeek
Llama 3.1 405B vs o1
OpenAI
Llama 3.1 405B vs Gemini 2.0 Flash
Google
Llama 3.1 405B vs DeepSeek V3
DeepSeek
Llama 3.1 405B vs Claude Sonnet 4.6
Anthropic

Frequently Asked Questions

How much does Llama 3.1 405B cost?

Llama 3.1 405B costs $2.70/1M per 1M input tokens and $2.70/1M per 1M output tokens.

What is Llama 3.1 405B's context window?

Llama 3.1 405B has a context window of 131K tokens, which means it can process up to 131,072 tokens in a single request.

Is Llama 3.1 405B open source?

Yes, Llama 3.1 405B is open source. The model weights are publicly available and can be self-hosted.

What is Llama 3.1 405B best used for?

Llama 3.1 405B is best suited for: coding, research, translation, reasoning. Open source — can be self-hosted for data privacy.

What is Llama 3.1 405B's ELO score?

Llama 3.1 405B has an ELO score of 1267 on the LMSYS Chatbot Arena leaderboard, placing it in the mid-tier range.