Amazon Nova Micro API Pricing: Industry-Lowest Cloud Token Economics
Comprehensive Amazon Nova Micro API pricing analysis ($0.035/M input, $0.14/M output), AWS Bedrock serverless speed, sub-100ms classification, and batch extraction ROI.
Full specs, context window and API limits →How much does Amazon Nova Micro cost per million tokens?
Amazon Nova Micro costs $0.035 per million input tokens and $0.14 per million output tokens ($0.06125/M blended at 3:1). The most economical text model available on AWS Bedrock, engineered for ultra-fast classification, routing, and high-volume data normalization. Verified 2026-09-08.
How much does Amazon Nova Micro cost per 1,000 requests?
Computed from generated token pricing. Each row assumes the listed input and output tokens per request; output is adjusted by this model's measured 0.76× verbosity factor.
| Request shape | Input tokens | Output tokens | Cost / 1,000 requests |
|---|---|---|---|
| Short | 100 | 50 | $0.0088 |
| Medium | 1,000 | 500 | $0.0882 |
| Long | 4,000 | 2,000 | $0.3528 |
Formula: ((input price × input tokens) + (output price × output tokens × verbosity factor)) ÷ 1,000,000 × 1,000. Assumptions: short 100/50, medium 1,000/500, long 4,000/2,000 input/output tokens per request. Verbosity run: 2026-06-16T20:31:30.728Z.
Three model-specific pricing decisions
Nova Micro owns text-only Bedrock economics. Vision-shaped work is ineligible for this unit-safe bill, not priced at zero.
1. Classification, routing, and short-generation ladder
| Workload | Input / output | 100K | 1M | 10M |
|---|---|---|---|---|
| Classification | 300 / 30 | $1.47 | $14.70 | $147.00 |
| Routing | 500 / 80 | $2.87 | $28.70 | $287.00 |
| Short generation | 1,000 / 500 | $10.50 | $105.00 | $1050.00 |
2. Micro-to-Lite accepted-result threshold
| Fixed input / output | Amazon Nova Micro | Amazon Nova Lite | Narrow decision boundary |
|---|---|---|---|
| Text classification · 300 / 30 | $1.47 | $2.52 | 10% accepted-result uplift required |
| Text generation · 1,000 / 500 | $10.50 | $18.00 | 15% accepted-result uplift required |
Formula: requests × (input tokens × input $/M + output tokens × output $/M) ÷ 1,000,000. The uplift is a planning threshold, not a measured quality claim.
Vision-shaped work: Ineligible — image units are not compatible with this text-only comparison and are not assigned a zero price.
3. Price / speed / AWS deployment matrix
| Dimension | Dated evidence | Safe boundary |
|---|---|---|
| Price | 2026-06-14 · $0.03 / $0.14 per M | On-demand listed text units |
| Speed | 168 tokens/sec; TTFT 220 ms; 5 measured samples | Measured/estimated state is visible |
| Region / quota / batch / cache / provisioned throughput / IAM | Unavailable | Do not infer a rate or adoption cost |
Verified 2026-06-14. Luna is the data owner. “Unavailable” means no compatible dated evidence was found; it is never treated as zero. First-party price source · Run this Batch 5 scenario.
All three Batch 5 decisions are server-rendered for Amazon Nova Micro; fixed inputs, formulas, dated sources, speed sample state, and unavailable mechanics are visible.
Batch 67 · exact model pricing decision contributions · verified 2026-09-07
Exact model boundary: Amazon nova-micro (slug nova-micro). First-party provider pricing and API documentation remain fact owners.
Ultra-budget token pricing and high-volume spend matrix
Frozen Batch 67 scenario board. Formula / deterministic rule: monthly_spend = calls * ((in_tokens * 0.03 + out_tokens * 0.14) / 1M) Boundary: Owns Nova Micro unit token economics and high-volume classification spend.
| Frozen scenario / field ID | Model, identity, provider, and evidence fields | Result | State |
|---|---|---|---|
batch67-nova-micro-m1-r1100K intent routing classifications | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=100K intent routing classifications; workload; prompt tokens; completion tokens; standard spend; batch spend; cost winner; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 100K intent routing classifications is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m1-r2500K customer inquiry triage calls | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=500K customer inquiry triage calls; workload; prompt tokens; completion tokens; standard spend; batch spend; cost winner; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 500K customer inquiry triage calls is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m1-r32M high-volume automated filter events | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=2M high-volume automated filter events; workload; prompt tokens; completion tokens; standard spend; batch spend; cost winner; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 2M high-volume automated filter events is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m1-r4batch API queue (50% discount) | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=batch API queue (50% discount); workload; prompt tokens; completion tokens; standard spend; batch spend; cost winner; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — batch API queue (50% discount) is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m1-r5provisioned throughput commitment | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=provisioned throughput commitment; workload; prompt tokens; completion tokens; standard spend; batch spend; cost winner; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — provisioned throughput commitment is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m1-r6unresolved billing currency | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=unresolved billing currency; workload; prompt tokens; completion tokens; standard spend; batch spend; cost winner; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — unresolved billing currency has no matched, dated bilateral observation. | FAIL CLOSED — manual, probe, or source evidence required |
First-party provenance: AWS Bedrock Nova pricing; verification date 2026-09-07. Missing or conflicting joins fail closed.
Sub-second latency SLA and intent routing turnaround audit
Frozen Batch 67 scenario board. Formula / deterministic rule: turnaround = ttft + (tokens_out / tps); Nova Micro optimized for lowest latency Boundary: Owns turnaround time benchmarks and user experience responsiveness.
| Frozen scenario / field ID | Model, identity, provider, and evidence fields | Result | State |
|---|---|---|---|
batch67-nova-micro-m2-r1real-time intent classification (<100ms) | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=real-time intent classification (<100ms); task; response SLA; TTFT; turnaround time; SLA compliance; responsiveness champion; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — real-time intent classification (<100ms) is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m2-r2customer support routing ticket (<200ms) | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=customer support routing ticket (<200ms); task; response SLA; TTFT; turnaround time; SLA compliance; responsiveness champion; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — customer support routing ticket (<200ms) is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m2-r3keyword extraction pass (<300ms) | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=keyword extraction pass (<300ms); task; response SLA; TTFT; turnaround time; SLA compliance; responsiveness champion; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — keyword extraction pass (<300ms) is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m2-r4high-concurrency request surge | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=high-concurrency request surge; task; response SLA; TTFT; turnaround time; SLA compliance; responsiveness champion; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — high-concurrency request surge is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m2-r5network contention latency buffer | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=network contention latency buffer; task; response SLA; TTFT; turnaround time; SLA compliance; responsiveness champion; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — network contention latency buffer is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m2-r6unmeasured speed fixture | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=unmeasured speed fixture; task; response SLA; TTFT; turnaround time; SLA compliance; responsiveness champion; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — unmeasured speed fixture is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
First-party provenance: Amazon Nova documentation; verification date 2026-09-07. Missing or conflicting joins fail closed.
Two-tier escalation architecture and routing cost savings
Frozen Batch 67 scenario board. Formula / deterministic rule: blended_cost = micro_volume * micro_cost + escalated_volume * escalated_cost Boundary: Owns two-tier architectural routing between ultra-budget triage and complex reasoning.
| Frozen scenario / field ID | Model, identity, provider, and evidence fields | Result | State |
|---|---|---|---|
batch67-nova-micro-m3-r1100% Nova Micro baseline ($0.03/$0.14) | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=100% Nova Micro baseline ($0.03/$0.14); escalation mix; total monthly queries; blended spend; cost savings vs pure Pro; accuracy retention; routing recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 100% Nova Micro baseline ($0.03/$0.14) is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m3-r290% Micro triage / 10% Nova Lite escalation | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=90% Micro triage / 10% Nova Lite escalation; escalation mix; total monthly queries; blended spend; cost savings vs pure Pro; accuracy retention; routing recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 90% Micro triage / 10% Nova Lite escalation is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m3-r380% Micro triage / 20% Nova Pro escalation | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=80% Micro triage / 20% Nova Pro escalation; escalation mix; total monthly queries; blended spend; cost savings vs pure Pro; accuracy retention; routing recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 80% Micro triage / 20% Nova Pro escalation is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m3-r450% Micro triage / 50% Nova Pro escalation | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=50% Micro triage / 50% Nova Pro escalation; escalation mix; total monthly queries; blended spend; cost savings vs pure Pro; accuracy retention; routing recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 50% Micro triage / 50% Nova Pro escalation is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m3-r5100% direct Nova Pro execution | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=100% direct Nova Pro execution; escalation mix; total monthly queries; blended spend; cost savings vs pure Pro; accuracy retention; routing recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 100% direct Nova Pro execution is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch67-nova-micro-m3-r6unresolved confidence threshold trigger | model=nova-micro; slug=nova-micro; provider=Amazon; scenario=unresolved confidence threshold trigger; escalation mix; total monthly queries; blended spend; cost savings vs pure Pro; accuracy retention; routing recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — unresolved confidence threshold trigger has no matched, dated bilateral observation. | FAIL CLOSED — manual, probe, or source evidence required |
First-party provenance: AWS Bedrock Nova pricing; verification date 2026-09-07. Missing or conflicting joins fail closed.
Method and limitations: formulas are deterministic; observed and assumed inputs are labeled; no missing provider, host, account, region, realm, alias, snapshot, revision, weight, artifact, control, tool, modality, workload, rate-period, timestamp, or result is transferred. Run the nova-micro Batch 67 scenario →
nova-microAmazon Nova Micro API Pricing: Industry-Lowest Cloud Token Economics
Amazon Nova Micro costs $0.035 per million input tokens and $0.14 per million output tokens ($0.06125/M blended at 3:1). The most economical text model available on AWS Bedrock, engineered for ultra-fast classification, routing, and high-volume data normalization. Verified 2026-09-08.
Amazon Nova Micro delivers world-class cost efficiency at approximately $0.06/M blended tokens.
| Scenario | Rendered Evidence & Bounds |
|---|---|
| Scenario 1 | Support ticket intent categorization (500 in, 40 out): $0.000023 per ticket |
| Scenario 2 | Customer feedback sentiment scoring (1K in, 30 out): $0.000039 per review |
| Scenario 3 | High-volume webhook data normalization (800 in, 50 out): $0.000035 per webhook |
| Scenario 4 | Real-time text autocomplete keystroke completion (200 in, 20 out): $0.000010 per keystroke |
| Scenario 5 | Log anomaly classification turn (1.5K in, 60 out): $0.000061 per log entry |
| Scenario 6 | Monthly 100M token classification fleet: $6.13 total API infrastructure spend |
Direct S3 connectivity and microscopic token prices make it the ultimate data warehouse ETL engine.
| Scenario | Rendered Evidence & Bounds |
|---|---|
| Scenario 1 | Processes 100,000 customer survey responses for less than $0.50 total API cost |
| Scenario 2 | Enriches 1 million database customer records for approximately $3.50 total spend |
| Scenario 3 | Direct integration with Amazon S3 and AWS Lambda eliminates middleware server hosting |
| Scenario 4 | High-throughput serverless concurrency handles burst traffic without provisioning servers |
| Scenario 5 | Strict JSON mode output ensures reliable ingestion into downstream PostgreSQL/DynamoDB |
| Scenario 6 | Cuts data warehouse ETL preparation expenditures by more than 90% compared to legacy pipelines |
Tiered Bedrock architecture slashes overall operational AI spend by 86% while boosting speed.
| Scenario | Rendered Evidence & Bounds |
|---|---|
| Scenario 1 | 1M queries routed via tiered architecture: $195.13 vs $1,400.00 monolithic Nova Pro fleet |
| Scenario 2 | Nova Micro handles 90% routine intent routing, sentiment tagging, and FAQ filtering |
| Scenario 3 | Nova Pro ($0.80/$3.20) reserved for complex multi-page document synthesis and vision |
| Scenario 4 | Overall AWS Bedrock API invoice reduced by 86.1% with zero loss in answer precision |
| Scenario 5 | Fleet average latency drops by 70% due to Nova Micro sub-80ms first-token generation |
| Scenario 6 | Standardized Bedrock Converse API parameters ensure frictionless handoff between models |
How fast is Amazon Nova Micro?
How much does Amazon Nova Micro cost at scale?
| Tokens / month | Est. cost (blended 3:1) |
|---|---|
| 100,000 | $0.0061 |
| 1,000,000 | $0.06 |
| 10,000,000 | $0.61 |
| 100,000,000 | $6.12 |
How does Amazon Nova Micro compare with other models?
What is Amazon Nova Micro best for?
What should you explore next for Amazon Nova Micro?
What are common questions about Amazon Nova Micro?
Is Amazon Nova Micro cheaper than Amazon Nova Lite?
Amazon Nova Micro costs $0.06/M blended tokens, Amazon Nova Lite costs $0.11/M — Amazon Nova Micro is cheaper.
How much does 1 million tokens cost with Amazon Nova Micro?
At a 3:1 input:output ratio, 1 million blended tokens costs approximately $0.06. Pure input costs $0.03/M; pure output costs $0.14/M.
What does Amazon Nova Micro cost at high volume?
At 100 million blended tokens a month, Amazon Nova Micro costs approximately $6.12. See the cost-at-scale table below for other volumes.