# Llama 3.3 70B

> Meta's Llama 3.3 open model, free on Groq, Cerebras and SambaNova within their limits.

- Maker: Meta (https://www.llama.com)
- Type: Chat, open weights
- Artificial Analysis Intelligence Index: Not scored yet
- Context window: 128K
- Output speed: 90 tokens a second
- Scores checked 30 October 2026 at https://artificialanalysis.ai/leaderboards/models

## Where to use it through an API

- Groq: model id `llama-3.3-70b-versatile` (free key available)
- Together AI: model id `meta-llama/Llama-3.3-70B-Instruct-Turbo`
- Cerebras: model id `llama-3.3-70b`
- Meta: model id `Llama-3.3-70B-Instruct`
- SambaNova: model id `Meta-Llama-3.3-70B-Instruct` (free key available)

## Use it free

- Groq free API key: how to get one: https://www.multiscreenai.com/free/groq-api-key
- SambaNova free API key: how to get one: https://www.multiscreenai.com/free/sambanova-api-key

## Use it with every other AI at once

MultiScreenAI (https://www.multiscreenai.com) puts Llama 3.3 70B on one screen next to ChatGPT, Claude, Gemini and 300+ models: one message goes to every panel, and a timed, blind test mode shows which answers best. Free during the beta with your own API keys.
AI agents and coding harnesses (Claude Code, Codex, Cursor, Gemini CLI) can ask Llama 3.3 70B and other models at once through the MCP server at https://www.multiscreenai.com/api/mcp/ai; setup: https://www.multiscreenai.com/mcp