All models / Meta
Llama 3.3 70B
Open-weights model under Llama 3.3 Community License. Pricing shown is a representative inference-provider rate (Together / Fireworks / Groq vary). Self-host for free.
At a glance
- Provider
- Meta
- Family
- llama-3
- Released
- 2024-12
- Status
- Active
- Context window
- 128K
- Max output
- 8K
- Knowledge cutoff
- 2023-12
- API id
- meta-llama/Llama-3.3-70B-Instruct
Pricing
USD per 1M tokens.
- Input
- $0.230
- Output
- $0.400
- Cache read
- —
- Cache write (5m)
- —
- Cache write (1h)
- —
- Batch input
- —
- Batch output
- —
Capabilities
- Tool use
- Yes
- Structured outputs
- Yes
- Vision
- No
- Extended thinking
- No
- Prompt caching
- No
- Batch API
- No
- Fine-tuning
- Yes
- Open weights
- Yes
Modalities
- Input
- text
- Output
- text
Compare with
Links
Last verified 2026-04-26. License: llama-community.