Llama 3.3 70B Instruct (free)
META-LLAMA Developer Architecture Profile
Intelligence (ELO)1455Chatbot Arena Verified
Max Context65,536Tokens
API Cost / 1MFREEBlended Prompt + Completion
Model Capabilities
- Classification
- Conversational
- Coding & Logic
- Fictional
The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out). The Llama 3.3 instruction tuned text only model...
Granular Pricing Matrix
Input Tokens (Prompt)Free / 1M
Output Tokens (Completion)Free / 1M
Pricing data via OpenRouter. Sync: 4/30/2026
Evaluate Competitors
VS Engine MatchupLlama 3.3 70B Instruct (free) vs Xiaomi: MiMo-V2.5-ProVS Engine MatchupLlama 3.3 70B Instruct (free) vs OpenAI: GPT-5.4 MiniVS Engine MatchupLlama 3.3 70B Instruct (free) vs Z.ai: GLM 5VS Engine MatchupLlama 3.3 70B Instruct (free) vs Mistral LargeVS Engine MatchupLlama 3.3 70B Instruct (free) vs Upstage: Solar Pro 3VS Engine MatchupLlama 3.3 70B Instruct (free) vs Anthropic: Claude Sonnet 4.5