Input / Prompt
$0.06/ 1M tokens
$0.0001 per 1K tokens
Output / Completion
$0.24/ 1M tokens
$0.0002 per 1K tokens
Prompt Cache Read
$0.01/ 1M tokens
Save 75% vs prompt
Batch Inference
$0.03/ 1M in
Save 50% via Batch API
Price Ratio
1 : 4.0
Prompt to completion factor
Regional Reach
3AWS Regions

Nova Sonic Spend Estimator

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

Workload Presets:
M tokens
M tokens
30%
Estimated Monthly Spend
$0.00
Fresh Prompts:$0.00
Cached Prompts:$0.00
Completions:$0.00
Active: Input: $0.06/1M ยท Output: $0.24/1M

Key Cost Levers & Architecture

Practical mechanisms to reduce inference costs with Nova Sonic on AWS
Prompt CachingSave ~75%

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

  • Read Rate: $0.01 / 1M tokens
  • Write Rate: $0.07 / 1M tokens
  • Ideal for: Multi-turn conversations, RAG document queries, structured schemas
Batch InferenceSave 50%

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

  • Batch Input: $0.03 / 1M tokens
  • Batch Output: $0.12 / 1M tokens
  • Ideal for: Offline classification, synthetic data generation, benchmark evals
Cross-Region & Tiers3 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 3 available AWS cloud regions
Baseline: $0.07/1M inGovCloud: $4.10/1M in3 Global Regions
Showing 2 benchmark regions
AWS RegionPrompt InputCompletion OutputCache ReadBatch InputPricing Tier
$3.40/1M$13.60/1Mโ€”/1Mโ€”/1M+4622% vs Baseline
Asia Pacific (Tokyo)ap-northeast-1
$0.07/1M$14.70/1Mโ€”/1Mโ€”/1MBaseline
EU (Stockholm)eu-north-1
$4.10/1M$0.26/1Mโ€”/1Mโ€”/1M+5594% vs Baseline

Developer Quickstart

Invoke Nova Sonic via Amazon Bedrock Converse API
Python (boto3) ยท Amazon Bedrock Runtime
import boto3

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

response = client.converse(
    modelId="amazon.nova-sonic-v1:0",
    messages=[
        {
            "role": "user",
            "content": [{"text": "Summarize key features and cost levers of Nova Sonic."}]
        }
    ],
    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")

Compare Nova Sonic with Alternatives

Open Multi-Model Estimator →
Nova Sonic vsVoxtral Mini 1.0Mistral AI
Input Rate:$0.02/1M
Output Rate:$0.02/1M
Compare Side-by-Side →
Nova Sonic vsVoxtral Small 1.0Mistral AI
Input Rate:$0.05/1M
Output Rate:$0.15/1M
Compare Side-by-Side →