DeepSeek V3.1
Amazon Bedrock token pricing and cost optimization reference for DeepSeek V3.1.Rates normalized to USD per 1 Million Tokens ($/1M) with prompt caching and batch API discounts across 13 AWS regions.
deepseek.deepseek-v3-1-v1:0On Google Cloud Vertex AI
The same model is sold through Vertex AI at one global rate per price type. These are Google's published rates, not the AWS rates above; both hosts are shown because the model is sold on both.
| Rate | Vertex AI, USD per 1M tokens |
|---|---|
| Input | $0.60 |
| Output | $1.70 |
| Cache read | $0.06 |
| Batch input | $0.30 |
| Batch output | $0.85 |
| Input, long context | — |
| Output, long context | — |
| Output, thinking | — |
Vertex AI rates are global: Google publishes one rate covering every region where the model runs, against the per-region AWS prices above. A dash means Google publishes no rate for that price type.
Alternatives to DeepSeek V3.1
| Option | Price vs DeepSeek V3.1 | Why look at it |
|---|---|---|
| DeepSeek v3.2Compare → | input+7%+$0.04/1M output+10%+$0.17/1M Standard tier, per 1M tokens | Next step up in price from DeepSeek |
| R1Compare → | input+133%+$0.77/1M output+221%+$3.72/1M Standard tier, per 1M tokens | Also from DeepSeek (flagship tier) |
DeepSeek V3.1 Spend Estimator
Model inference spend by prompt volume, cache hit rate, and batch processing.
Key Cost Levers & Architecture
Practical mechanisms to reduce inference costs with DeepSeek V3.1 on AWSCaches repetitive system instructions, schemas, and document prefixes in memory with 5-minute and extended 1-hour TTL tiers.
- Read Rate: $0.14 / 1M tokens
- Write Rate (5m TTL): $0.72 / 1M tokens
- Ideal for: Multi-turn conversations, RAG document queries, structured schemas
Asynchronous bulk processing submitted via Amazon S3 with a standard 24-hour turnaround SLA.
- Batch Input: $0.29 / 1M tokens
- Batch Output: $0.84 / 1M tokens
- Ideal for: Offline classification, synthetic data generation, benchmark evals
AWS Bedrock supports standard on-demand inference, priority reserved throughput, and cross-region routing.
- Standard: Lowest latency, pay-per-token with zero commitments
- Cross-Region Routing: Automatically burst traffic to available regional capacity
- GovCloud: Dedicated isolated compliance partitions with +20% uplift
Rates by region
Live rates across all 13 available AWS cloud regions| AWS Region | Prompt Input $/1M | Completion Output $/1M | Cache Read $/1M | Batch Input $/1M | Pricing Tier |
|---|---|---|---|---|---|
| us-east-1 US East (N. Virginia) | $0.58 | $1.68 | — | $0.29 | Baseline |
| us-west-2 US West (Oregon) | $0.58 | $1.68 | — | $0.29 | Baseline |
| eu-central-1 EU (Frankfurt) | $0.70 | $2.02 | — | $0.35 | +20% vs Baseline |
| eu-west-1 EU (Ireland) | $0.70 | $2.02 | — | $0.35 | +20% vs Baseline |
| ap-northeast-1 Asia Pacific (Tokyo) | $0.70 | $2.03 | — | $0.35 | +21% vs Baseline |
| ap-southeast-3 Asia Pacific (Jakarta) | $0.60 | $1.74 | — | $0.30 | +3% vs Baseline |
| ap-south-1 Asia Pacific (Mumbai) | $0.68 | $1.98 | — | $0.34 | +18% vs Baseline |
| ap-southeast-2 Asia Pacific (Sydney) | $0.60 | $1.73 | — | $0.30 | +3% vs Baseline |
| eu-west-2 EU (London) | $0.90 | $2.61 | — | $0.45 | +55% vs Baseline |
| eu-south-1 EU (Milan) | $0.70 | $2.02 | — | $0.35 | +20% vs Baseline |
| eu-north-1 EU (Stockholm) | $0.58 | $1.68 | — | $0.29 | Baseline |
| sa-east-1 South America (Sao Paulo) | $0.70 | $2.02 | — | $0.35 | +20% vs Baseline |
| us-east-2 US East (Ohio) | $0.58 | $1.68 | — | $0.29 | Baseline |
Developer Quickstart
Invoke DeepSeek V3.1 via Amazon Bedrock Converse APIimport boto3
# Amazon Bedrock Converse API invocation
client = boto3.client("bedrock-runtime", region_name="us-east-1")
response = client.converse(
modelId="deepseek.deepseek-v3-1-v1:0",
messages=[
{
"role": "user",
"content": [{"text": "Summarize key features and cost levers of DeepSeek V3.1."}]
}
],
inferenceConfig={
"maxTokens": 1024,
"temperature": 0.7
}
)
output_text = response["output"]["message"]["content"][0]["text"]
usage = response["usage"]
print(f"Response: {output_text}")
print(f"Usage: {usage['inputTokens']} in, {usage['outputTokens']} out")