Amazon Bedrock pricing, checked at the source
On-demand prices for the models on Amazon Bedrock in US East (N. Virginia), read off the AWS pricing page with a link on every row. Estimate a monthly bill below.
Amazon Bedrock prices per 1M tokens
Standard prices in USD. Each row links to Amazon Web Services's pricing page and shows the day we checked it.
| Model | Input | Cached input | Output | Batch inference (in / out) | Notes | Source |
|---|---|---|---|---|---|---|
| Claude Opus 5.5us-east-1, in-region | $4.40 | $0.22 | $22.00 | none listed | Global cross-region: $4 in, $0.20 cache read, $20 out. 5m cache write $5.50. | Sourcechecked 2026-10-25 |
| Claude Sonnet 5.5us-east-1, in-region | $2.20 | $0.11 | $11.00 | none listed | Global cross-region: $2 in, $0.10 cache read, $10 out. Page says cache read price was cut 50% from October 7, 2026. | Sourcechecked 2026-10-25 |
| Claude Haiku 5.5us-east-1, in-region | $0.11 | $0.011 | $0.55 | none listed | Global cross-region: $0.10 in, $0.01 cache read, $0.50 out. Long-context variant is priced separately. | Sourcechecked 2026-10-25 |
| Amazon Nova 2 Liteus-east-1, in-region | $0.33 | $0.0825 | $2.75 | none listed | Cached price worked out from the page rule that cache reads are 75% less than input. Global cross-region: $0.30 in, $2.50 out. | Sourcechecked 2026-10-25 |
| Amazon Nova Microus-east-1, in-region | $0.035 | $0.00875 | $0.14 | $0.0175 / $0.07 | Batch: $0.0175 in, $0.07 out. | Sourcechecked 2026-10-25 |
| Llama 4 Maverick 17Bus-east-1, in-region | $0.24 | none listed | $0.97 | $0.12 / $0.485 | Batch: $0.12 in, $0.485 out. | Sourcechecked 2026-10-25 |
| DeepSeek v3.2us-east-1, in-region | $0.62 | none listed | $1.85 | none listed | Sourcechecked 2026-10-25 | |
| Mistral Large 3us-east-1, in-region | $0.50 | none listed | $1.50 | none listed | Sourcechecked 2026-10-25 | |
| gpt-oss-120bus-east-1, in-region | $0.15 | none listed | $0.60 | $0.075 / $0.30 | Batch and Flex: $0.075 in, $0.30 out. | Sourcechecked 2026-10-25 |
| MiniMax M2.5us-east-1, in-region | $0.30 | none listed | $1.20 | none listed | Sourcechecked 2026-10-25 |
Terms on this page: Token · Prompt caching · Context window · Batch API
Estimate your Amazon Web Services bill
Enter your monthly tokens and how they split between new input, cached input and output. The result updates as you type.
Write 40k, 250M or 5B. Input, cached input and output tokens together.
100M tokens a month costs from $2.50 (Amazon Nova Micro) to $285.56 (Claude Opus 5.5).
| Model | Per month | Per 1M tokens at this split |
|---|---|---|
| Amazon Nova Microus-east-1, in-region | $2.50 | $0.025025 |
| Claude Haiku 5.5us-east-1, in-region | $7.46 | $0.07458 |
| gpt-oss-120bus-east-1, in-region | $17.25 | $0.1725 |
| Llama 4 Maverick 17Bus-east-1, in-region | $27.65 | $0.2765 |
| Amazon Nova 2 Liteus-east-1, in-region | $30.75 | $0.30745 |
| MiniMax M2.5us-east-1, in-region | $34.50 | $0.345 |
| Mistral Large 3us-east-1, in-region | $55.00 | $0.55 |
| DeepSeek v3.2us-east-1, in-region | $68.15 | $0.6815 |
| Claude Sonnet 5.5us-east-1, in-region | $142.78 | $1.4278 |
| Claude Opus 5.5us-east-1, in-region | $285.56 | $2.8556 |
Standard prices in USD on 2026-10-25, before tax. Long-prompt tiers are left out; see the price table.
Amazon Web Services embeddings prices
Prices per 1M input tokens. Embedding models have no output price.
| Model | Price as listed | Per 1M tokens | Notes | Source |
|---|---|---|---|---|
| Amazon Titan Text Embeddings V2us-east-1 on-demand | $0.02 per 1M tokens | $0.02 | Batch: $0.01 per 1M input tokens. | Sourcechecked 2026-10-25 |
| Embed 3 English (Cohere)us-east-1 on-demand | $0.10 per 1M tokens | $0.10 | Image input is $0.0001 per image. | Sourcechecked 2026-10-25 |
| Embed 4 Model (Cohere)us-east-1 on-demand | $0.12 per 1M tokens | $0.12 | Sourcechecked 2026-10-25 | |
| Amazon Nova Multimodal Embeddingsus-east-1 on-demand, text input | $0.135 per 1M tokens | $0.135 | Batch: $0.0675 per 1M text tokens. | Sourcechecked 2026-10-25 |
Write 40k, 250M or 5B.
100M tokens a month costs from $2.00 (Amazon Titan Text Embeddings V2) to $13.50 (Amazon Nova Multimodal Embeddings).
| Model | Per month | Per 1M tokens |
|---|---|---|
| Amazon Titan Text Embeddings V2us-east-1 on-demand | $2.00 | $0.02 |
| Embed 3 English (Cohere)us-east-1 on-demand | $10.00 | $0.10 |
| Embed 4 Model (Cohere)us-east-1 on-demand | $12.00 | $0.12 |
| Amazon Nova Multimodal Embeddingsus-east-1 on-demand, text input | $13.50 | $0.135 |
List prices in USD on 2026-10-25, before tax, free tiers and add-ons. Rows billed another way are in the price table only.
Amazon Web Services image generation prices
Prices per image where the vendor lists one. Rows billed per megapixel or per token show how they bill instead.
| Model | Price as listed | Per image | Notes | Source |
|---|---|---|---|---|
| Amazon Titan Image Generator v2us-east-1 on-demand | $0.01 per image | $0.01 | Larger image size, Standard quality. Premium is $0.012. Smaller than 512 x 512 is $0.008. | Sourcechecked 2026-10-25 |
| Amazon Nova Canvasus-east-1 on-demand | $0.04 per image | $0.04 | Up to 1024 x 1024, Standard quality. Premium is $0.06. | Sourcechecked 2026-10-25 |
| Amazon Nova Canvasus-east-1 on-demand | $0.06 per image | $0.06 | Up to 2048 x 2048, Standard quality. Premium is $0.08. | Sourcechecked 2026-10-25 |
Write 500, 10k or 1M.
1k images a month costs from $10.00 (Amazon Titan Image Generator v2) to $60.00 (Amazon Nova Canvas).
| Model | Per month | Per image |
|---|---|---|
| Amazon Titan Image Generator v2us-east-1 on-demand | $10.00 | $0.01 |
| Amazon Nova Canvasus-east-1 on-demand | $40.00 | $0.04 |
| Amazon Nova Canvasus-east-1 on-demand | $60.00 | $0.06 |
List prices in USD on 2026-10-25, before tax, free tiers and add-ons. Rows billed another way are in the price table only.
How the bill is worked out
Batch is half price
Global routing is cheaper
Service tiers
Cache reads on Nova
Amazon Bedrock pricing questions
How much does the Amazon Bedrock cost?
Amazon Web Services bills its language models per million tokens, with separate prices for input, cached input and output. On 2026-10-25, standard prices ran from $0.035 (Amazon Nova Micro) to $4.40 (Claude Opus 5.5) per 1M input tokens, and $0.14 (Amazon Nova Micro) to $22.00 (Claude Opus 5.5) per 1M output tokens. Embeddings runs from $0.02 (Amazon Titan Text Embeddings V2) to $0.135 (Amazon Nova Multimodal Embeddings) per 1M tokens. Image generation runs from $0.01 (Amazon Titan Image Generator v2) to $0.06 (Amazon Nova Canvas) per image.
Does Amazon Web Services offer a batch discount?
On some models. Batch inference prices are listed for Amazon Nova Micro, Llama 4 Maverick 17B, gpt-oss-120b. The price table shows which rows have one.
How does prompt caching change a Amazon Web Services bill?
For Amazon Nova models, cache read input tokens cost 75% less than the normal input price.
Where do these prices come from?
From Amazon Web Services's own pricing page, https://aws.amazon.com/bedrock/pricing/, checked on 2026-10-25. Every row links to it. When a price cannot be confirmed, the row says to check the vendor's page instead of showing a number.
Is Claude on Bedrock dearer than Anthropic's own API?
In region, a little. Bedrock's global cross-region price for Claude Opus 5.5 is $4 input and $20 output per 1M tokens, the same as Anthropic's own list price. The in-region us-east-1 price in the table is about 10% higher.
Compare and plan
Building with the Amazon Bedrock? OceanLab builds AI into business workflows, with people kept in charge. AI automation →
