DeepSeek v3.2 Pricing & Token Calculator (DeepSeek)

Input / Prompt

$0.31/ 1M tokens

$0.0003 per 1K tokens

Output / Completion

$0.93/ 1M tokens

$0.0009 per 1K tokens

Prompt Cache Read

$0.08/ 1M tokens

Save 75% vs prompt

Batch Inference

$0.31/ 1M in

Save 0% via Batch API

Price Ratio

1 : 3.0

Prompt to completion factor

Regional Reach

13AWS Regions

Compare 13 regions ↓

DeepSeek v3.2 Spend Estimator

Model inference spend by prompt volume, cache hit rate, and batch processing.

Prompt Tokens (Millions / month)

M tokens

Completion Tokens (Millions / month)

M tokens

Prompt Cache Hit Rate(75% discount on cached context)

30%

Batch API Mode50% discount on non-realtime async jobs via S3

Estimated Monthly Spend

$0.00

Fresh Prompts:$0.00

Cached Prompts:$0.00

Completions:$0.00

Active: Input: $0.31/1M · Output: $0.93/1M

Key Cost Levers & Architecture

Practical mechanisms to reduce inference costs with DeepSeek v3.2 on AWS

Prompt CachingSave ~75%

Caches repetitive system instructions, schemas, and document prefixes in memory with a 5-minute TTL.

Read Rate: $0.08 / 1M tokens
Write Rate: $0.39 / 1M tokens
Ideal for: Multi-turn conversations, RAG document queries, structured schemas

Batch InferenceSave 0%

Asynchronous bulk processing submitted via Amazon S3 with a standard 24-hour turnaround SLA.

Batch Input: $0.31 / 1M tokens
Batch Output: $0.93 / 1M tokens
Ideal for: Offline classification, synthetic data generation, benchmark evals

Cross-Region & Tiers13 Regions

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

Regional Pricing Matrix

Live rates across all 13 available AWS cloud regions

Baseline: $0.31/1M inGovCloud: $1.29/1M in13 Global Regions

AWS Region	Prompt Input	Completion Output	Cache Read	Batch Input	Pricing Tier
US East (N. Virginia)`us-east-1`	$0.31/1M	$0.93/1M	—/1M	$0.31/1M	Baseline
US West (Oregon)`us-west-2`	$1.08/1M	$0.93/1M	—/1M	$0.31/1M	+250% vs Baseline
EU (Frankfurt)`eu-central-1`	$0.74/1M	$2.22/1M	—/1M	$0.37/1M	+139% vs Baseline
EU (Ireland)`eu-west-1`	$0.74/1M	$1.11/1M	—/1M	$0.37/1M	+139% vs Baseline
Asia Pacific (Tokyo)`ap-northeast-1`	$0.37/1M	$2.22/1M	—/1M	$0.37/1M	+19% vs Baseline
Asia Pacific (Jakarta)`ap-southeast-3`	$0.37/1M	$2.22/1M	—/1M	$0.37/1M	+19% vs Baseline
Asia Pacific (Mumbai)`ap-south-1`	$0.37/1M	$2.22/1M	—/1M	$0.37/1M	+19% vs Baseline
Asia Pacific (Sydney)`ap-southeast-2`	$1.12/1M	$1.91/1M	—/1M	$0.32/1M	+261% vs Baseline
EU (London)`eu-west-2`	$0.48/1M	$1.43/1M	—/1M	$0.48/1M	+55% vs Baseline
EU (Milan)`eu-south-1`	$1.29/1M	$2.22/1M	—/1M	$0.37/1M	+318% vs Baseline
EU (Stockholm)`eu-north-1`	$0.74/1M	$1.11/1M	—/1M	$0.37/1M	+139% vs Baseline
South America (Sao Paulo)`sa-east-1`	$0.74/1M	$1.11/1M	—/1M	$0.37/1M	+139% vs Baseline
US East (Ohio)`us-east-2`	$0.31/1M	$0.93/1M	—/1M	$0.31/1M	Baseline

Developer Quickstart

Invoke DeepSeek v3.2 via Amazon Bedrock Converse API

Python (boto3) · Amazon Bedrock Runtime

import boto3

# Amazon Bedrock Converse API invocation
client = boto3.client("bedrock-runtime", region_name="us-east-1")

response = client.converse(
    modelId="deepseek.deepseek-v3-2-v1:0",
    messages=[
        {
            "role": "user",
            "content": [{"text": "Summarize key features and cost levers of DeepSeek v3.2."}]
        }
    ],
    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")