Llama 3.3 70B pricing
Llama 3.3 70B by Meta is sold on Amazon Bedrock and Microsoft Foundry (Azure). It costs $0.72 per million input tokens and $0.72 per million output tokens on Amazon Bedrock (in us-east-1); $0.71 per million input tokens and $0.71 per million output tokens on Microsoft Foundry (Azure) (global deployment).
Price on each platform
| Platform | Input | Output | Cached input | Batch input / output |
|---|---|---|---|---|
| Amazon Bedrock | $0.72 | $0.72 | – | $0.36 / $0.36 |
| Microsoft Foundry (Azure) | $0.71 | $0.71 | – | – |
The lowest standard price on each platform, in USD per 1M tokens.
Amazon Bedrock
| Tier | Runs in | Input | Output |
|---|---|---|---|
| Standard | In region (us-east-1, us-east-2, us-west-2) | $0.72 | $0.72 |
| Batch | In region (us-east-1, us-east-2, us-west-2) | $0.36 | $0.36 |
Microsoft Foundry (Azure)
| Tier | Runs in | Input | Output |
|---|---|---|---|
| Standard | Global (40 regions) | $0.71 | $0.71 |
| Standard | Geography (data zone) (22 regions) | $0.781 | $0.781 |
| Standard | In region (9 regions) | $0.781 | $0.781 |
| Standard | In region (6 regions) | $0.859 | $0.859 |
| Standard | In region (southeastasia, westeurope) | $0.937 | $0.937 |
More Meta models
Llama 3 8B · Llama 3 70B · Llama 3.1 8B · Llama 3.1 70B · Llama 3.1 70B Latency Optimized · Llama 3.1 405B · Llama 3.1 405B Latency Optimized · Llama 3.2 1B · Llama 3.2 3B · Llama 3.2 11B · Llama 3.2 90B · Llama 4 Maverick 17B · Llama 4 Scout 17B · Llama2-13B · Llama2-70B
Prices in USD per 1 million tokens, from OpenAI and Anthropic's pricing pages, the AWS Price List API (Bedrock) and the Azure Retail Prices API (Microsoft Foundry). Cloud prices are the lowest in any region; global deployments can run anywhere.
Prices as of 12 Nov 2026. All AI models