Foundation Model · Amazon BedrockLlama FamilyLightweight TierTextMulti-Tier Pricing

Llama 3.2 3B

Amazon Bedrock token pricing and cost optimization reference for Llama 3.2 3B.Rates normalized to USD per 1 Million Tokens ($/1M) with prompt caching and batch API discounts across 6 AWS regions.

ProviderMeta

Maker HQMenlo Park, California, USA

Class / TierLightweight

ModalityText

Prompt CachingSave 75%

Batch APISave 50%

AWS Regions6 Global

Bedrock Model IDmeta.llama-3-2-3b-v1:0

Input / Prompt

$0.15/ 1M tokens

$0.0001 per 1K tokens

Output / Completion

$0.15/ 1M tokens

$0.0001 per 1K tokens

Prompt Cache Read

$0.04/ 1M tokens

Save 75% vs prompt

Batch Inference

$0.07/ 1M in

Save 50% via Batch API

Price Ratio

1 : 1.0

Prompt to completion factor

Regional Reach

6AWS Regions

Compare 6 regions ↓

Alternatives to Llama 3.2 3B

Option	Price vs Llama 3.2 3B	Why look at it
Llama 3.2 1B Compare →	input-33%-$0.05/1M output-33%-$0.05/1M Standard tier, per 1M tokens	Next step down in price from Meta
Llama 3.2 11B Compare →	input+7%+$0.01/1M output+7%+$0.01/1M Standard tier, per 1M tokens	Next step up in price from Meta
Llama 3.1 8B Compare →	input+47%+$0.07/1M output+47%+$0.07/1M Standard tier, per 1M tokens	Same Llama line, different version
Llama 4 Scout 17B Compare →	input+13%+$0.02/1M output+340%+$0.51/1M Standard tier, per 1M tokens	Also from Meta

Llama 3.2 3B 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: $0.15/1M · Output: $0.15/1M

Key Cost Levers & Architecture

Practical mechanisms to reduce inference costs with Llama 3.2 3B on AWS

Prompt CachingSave ~75%

Caches repetitive system instructions, schemas, and document prefixes in memory with 5-minute and extended 1-hour TTL tiers.

Read Rate: $0.04 / 1M tokens
Write Rate (5m TTL): $0.19 / 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: $0.07 / 1M tokens
Batch Output: $0.07 / 1M tokens
Ideal for: Offline classification, synthetic data generation, benchmark evals

Cross-Region & Tiers6 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 6 available AWS cloud regions

Baseline: $0.15/1M inGovCloud: $0.20/1M in6 Global Regions

AWS Region	Prompt Input $/1M	Completion Output $/1M	Cache Read $/1M	Batch Input $/1M	Pricing Tier
us-east-1 US East (N. Virginia)	$0.15	$0.15	—	—	Baseline
us-west-2 US West (Oregon)	$0.15	$0.15	—	$0.07	Baseline
eu-central-1 EU (Frankfurt)	$0.19	$0.19	—	$0.10	+27% vs Baseline
eu-west-1 EU (Ireland)	$0.17	$0.17	—	—	+13% vs Baseline
eu-west-3 EU (Paris)	$0.20	$0.20	—	—	+33% vs Baseline
us-east-2 US East (Ohio)	$0.15	$0.15	—	—	Baseline

Developer Quickstart

Invoke Llama 3.2 3B 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="meta.llama-3-2-3b-v1:0",
    messages=[
        {
            "role": "user",
            "content": [{"text": "Summarize key features and cost levers of Llama 3.2 3B."}]
        }
    ],
    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")