Home › Providers › Meta (Llama)
Meta (Llama) API pricing
Llama 4 and 3.3, open weights. 3 models priced per 1,000,000 tokens, verified November 2026. Cheapest: Llama 4 Scout at $0.18 input. Most expensive: Llama 3.3 70B at $0.88 output.
Meta (Llama) price table
| Model | Input | Cached | Cache write | Output | Context | Batch |
|---|---|---|---|---|---|---|
| Llama 4 Scout | $0.18 | $0.018 |  | $0.59 | 10,000,000 | Yes |
| Llama 4 Maverick | $0.27 | $0.027 |  | $0.85 | 1,000,000 | Yes |
| Llama 3.3 70B | $0.88 | $0.088 |  | $0.88 | 128,000 | Yes |
How Meta (Llama) compares with other providers
| Provider | Models | Cheapest input | Cheapest output | Flagship output |
|---|---|---|---|---|
| OpenAI | 13 | $0.10 | $0.50 | $50 |
| Anthropic | 5 | $1 | $5 | $50 |
| Google Gemini | 6 | $0.30 | $2.50 | $12 |
| xAI Grok | 2 | $1.25 | $2.50 | $6 |
| DeepSeek | 2 | $0.30 | $1.20 | $3.96 |
| Meta (Llama) | 3 | $0.18 | $0.59 | $0.85 |
| Alibaba (Qwen) | 3 | $0.05 | $0.20 | $2 |
| Mistral AI | 3 | $0.10 | $0.30 | $6 |
Workload cost on Meta (Llama)
| Model | Support chat 1K reqs · 4K/600 | RAG + cache 1K reqs · 50K/1K | Nightly batch 10K reqs · 10K/2K |
|---|---|---|---|
| Llama 4 Scout | $1.07 | $2.31 | $14.90 |
| Llama 4 Maverick | $1.59 | $3.43 | $22 |
| Llama 3.3 70B | $4.05 | $9.28 | $52.80 |
FAQ
What is the cheapest Meta (Llama) model?
Llama 4 Scout at $0.18 per 1M input and $0.59 per 1M output tokens.
What is the most expensive Meta (Llama) model?
Llama 3.3 70B at $0.88 per 1M input and $0.88 per 1M output tokens.
Does Meta (Llama) charge for prompt caching?
Yes. Cached reads are published per model; on this provider they range from $0.018 to $0.088 per 1M tokens.
Other providers
- OpenAI pricing
- Anthropic pricing
- Google Gemini pricing
- xAI Grok pricing
- DeepSeek pricing
- Alibaba (Qwen) pricing
- Mistral AI pricing
Calculate your monthly Meta (Llama) bill