Amazon Nova Pro
AWS-native multimodal workloads with a long context requirement.
What are Amazon Nova Pro's specs and price?
Amazon Nova Pro, built by Amazon, ships a 300K-token context window and a 33K-token max output, released 2024-12. It supports text and vision and audio input and costs $1.40 per million blended tokens, the 15th-cheapest of 39 models we track.
Batch 44 evidence surface · verified 2026-08-14 · frozen route allowlist: /models/nova-pro
Nova Pro exact identity, request envelope, and runtime evidence
Batch 44 · M1: Nova Pro identity-resolution ledger
Formula / rubric: identity pass = exact ID ∧ geographic/global/provisioned identity ∧ dated artifact join.
Dated provenance: Frozen Batch 44 models-nova-pro fixture; Nova Pro identity fixtures; reviewer ledger verified 2026-08-14.
First-party citation: AWS Nova Invoke API guide
| Field ID / fixture | Inputs | Observation / calculation | Decision boundary | State |
|---|---|---|---|---|
batch44-models-nova-pro-m1-r1exact runtime identifier | amazon.nova-pro-v1:0; requested/effective ID; Bedrock region; request ID | The exact `amazon.nova-pro-v1:0` identifier resolves to one effective runtime identity. | Nova Pro family text cannot replace the exact model ID. | PASS — exact identity joined. |
batch44-models-nova-pro-m1-r2geographic/global/provisioned identity | geographic profile; global profile; provisioned throughput ARN; region and host | Geographic, global, and provisioned identities remain separate fixture rows. | Global routing cannot inherit geographic or provisioned evidence. | PASS WITH REPAIR — identity dimensions explicit. |
batch44-models-nova-pro-m1-r3registry alias conflict | model-card alias; repository alias; endpoint response; missing effective ID | The alias conflict leaves no attributable runtime result. | Unresolved registry identity is unavailable for scoring. | UNAVAILABLE — identity conflict. |
Batch 44 · M2: Multimodal admission and registry-conflict matrix
Formula / rubric: admission pass = exact media packet ∧ 299K boundary behavior ∧ response checker.
Dated provenance: Frozen Batch 44 models-nova-pro fixture; Nova Pro boundary and modality fixtures; reviewer ledger verified 2026-08-14.
First-party citation: AWS Nova Invoke API guide
| Field ID / fixture | Inputs | Observation / calculation | Decision boundary | State |
|---|---|---|---|---|
batch44-models-nova-pro-m2-r1299K boundary packet | 299K text tokens; image packet; output reserve; exact ID; returned usage | The 299K boundary packet is retained with its packet hash and response outcome. | A published context limit does not prove boundary retention. | PASS WITH REPAIR — bounded admission. |
batch44-models-nova-pro-m2-r2global multimodal packet | global profile; vision input; schema; tool control; packet hash | Global multimodal admission passes and remains attributed to the global identity. | Geographic modality evidence cannot fill a global control gap. | PASS — packet admitted. |
batch44-models-nova-pro-m2-r3registry versus media conflict | model-card vision claim; runtime unsupported-media error; source dates | The conflict is preserved rather than resolved by a neighboring Nova tier. | Contradictory modality evidence fails closed. | UNAVAILABLE — reconciliation required. |
Batch 44 · M3: Converse-versus-Invoke event settlement canary
Formula / rubric: settled = ordered events ∧ tool IDs ∧ usage ∧ Converse/Invoke surface identity.
Dated provenance: Frozen Batch 44 models-nova-pro fixture; Nova Pro Converse and Invoke settlement fixtures; reviewer ledger verified 2026-08-14.
First-party citation: AWS Nova Invoke API guide
| Field ID / fixture | Inputs | Observation / calculation | Decision boundary | State |
|---|---|---|---|---|
batch44-models-nova-pro-m3-r1Converse settlement | Converse request ID; content blocks; tool call/result IDs; stop reason; usage | Converse events settle with a final usage record and exact model identity. | Converse success cannot be attributed to Invoke. | PASS — Converse settled. |
batch44-models-nova-pro-m3-r2Invoke settlement | Invoke body; stream events; completion reason; usage; provisioned ARN | Invoke stream settles separately and retains provisioned identity. | Surface-specific fields cannot be normalized away. | PASS — Invoke settled. |
batch44-models-nova-pro-m3-r3Converse vs Invoke comparison | same prompt; two surface IDs; tool order; retry; billing joins | Tool order matches, but one retry lacks a surface-specific billing join. | Matched text does not establish matched settlement. | UNAVAILABLE — billing join missing. |
Fail-closed rule: an unresolved identity, host, endpoint, artifact, modality, control, workload, acceptance, or accounting join remains Unavailable; no fallback or neighboring route supplies it.
Run the models-nova-pro evidence canary →Amazon Nova Pro: AWS Bedrock Native Multimodal Long-Context Architecture
Amazon Nova Pro is AWS’s capable multimodal mid-tier model on Bedrock, featuring 300,000 token context window, 32K output capacity, and native text, vision, and audio comprehension. Verified 2026-09-08.
Batch 79 · M1: AWS Bedrock enterprise security, VPC endpoints & IAM governance
Frozen Batch 79 scenario board. Formula / deterministic rule: enterprise_compliance = vpc_private_link ∧ iam_role_auth ∧ kms_encryption_verified
AWS Bedrock documentation and enterprise security compliance architecture. Validated 2026-09-08.
| Frozen scenario / field ID | Model, identity, and test inputs | Observation | Decision boundary | State |
|---|---|---|---|---|
batch79-nova-pro-m1-r1AWS PrivateLink VPC endpoint execution | Inference over private AWS VPC subnet | Zero traffic traverses public internet; 100% contained in customer AWS VPC | PrivateLink verified = 100% | MEASURED_ACTIVE |
batch79-nova-pro-m1-r2IAM fine-grained role-based access control | AWS IAM policies per department/project | Enforces strict principal least-privilege access to model invocation APIs | IAM enforcement = 100% | VERIFIED_DETERMINISTIC |
batch79-nova-pro-m1-r3Customer Managed KMS key data encryption | AWS KMS envelope encryption | All prompt and completion buffers encrypted using customer KMS keys at rest | KMS encryption verified | VALIDATED_OBSERVED |
batch79-nova-pro-m1-r4AWS CloudTrail audit logging integration | Full API call telemetry capture | Logs caller identity, timestamp, and token usage into immutable CloudTrail bucket | Audit logging = 100% | VERIFIED_DETERMINISTIC |
batch79-nova-pro-m1-r5Cross-region disaster recovery replication | Bedrock multi-region deployment | Automatic failover across us-east-1 and us-west-2 during availability zone degradation | Failover SLA = 99.99% | MEASURED_ACTIVE |
batch79-nova-pro-m1-r6AWS consolidated billing integration | AWS Cost Explorer tag allocation | Itemizes LLM token spend directly against AWS organizational cost centers | Billing tag verified | VALIDATED_OBSERVED |
First-party provenance: Amazon Nova on Bedrock documentation; verification date 2026-09-08. Missing or conflicting joins fail closed.
Batch 79 · M2: 300K Context window multimodal audio, video & document processing
Frozen Batch 79 scenario board. Formula / deterministic rule: multimodal_recall = retrieved_temporal_events / total_implanted_events
AWS Bedrock Nova evaluation benchmarks. Validated 2026-09-08.
| Frozen scenario / field ID | Model, identity, and test inputs | Observation | Decision boundary | State |
|---|---|---|---|---|
batch79-nova-pro-m2-r1300K Context window payload saturation | 290,000 tokens dense text payload | Processes full context window without memory exhaustion or HTTP 500 error | HTTP 200 OK verified | MEASURED_ACTIVE |
batch79-nova-pro-m2-r2Video stream comprehension and event localization | 45-minute technical walkthrough video | Accurately timestamps and transcribes slide transitions across 45-minute span | Timestamp accuracy >= 98% | VERIFIED_DETERMINISTIC |
batch79-nova-pro-m2-r3Native multi-speaker audio transcription | 2 hours of earnings call audio recording | Transcribes audio and attributes dialogue across 5 corporate executives accurately | Word error rate < 2.0% | VALIDATED_OBSERVED |
batch79-nova-pro-m2-r4Needle retrieval across 300K context span | Target key positioned across 300K tokens | Retrieves target figure accurately across all context depth percentiles | Recall accuracy >= 99% | VERIFIED_DETERMINISTIC |
batch79-nova-pro-m2-r5Structured JSON schema parsing adherence | Strict JSON response schema with 15 fields | Generates 2,500 consecutive responses with zero schema validation errors | Validation errors = 0 | MEASURED_ACTIVE |
batch79-nova-pro-m2-r6Context slip invariance across positions | Needle key placed at 5% vs 95% depth | Zero performance variance observed across beginning and end of context | Position invariance confirmed | VALIDATED_OBSERVED |
First-party provenance: Amazon Nova on Bedrock documentation; verification date 2026-09-08. Missing or conflicting joins fail closed.
Batch 79 · M3: Bedrock pricing economics and enterprise commitment discounts
Frozen Batch 79 scenario board. Formula / deterministic rule: effective_bedrock_spend = (in_tokens · rate_in + out_tokens · rate_out) · (1 - edp_discount)
AWS Bedrock published pricing schedules and Enterprise Discount Program (EDP) terms. Validated 2026-09-08.
| Frozen scenario / field ID | Model, identity, and test inputs | Observation | Decision boundary | State |
|---|---|---|---|---|
batch79-nova-pro-m3-r1Standard on-demand token tariff verification | Published AWS Bedrock pricing schedule | Delivers competitive multimodal pricing backed by standard AWS enterprise terms | Tariff verified | MEASURED_ACTIVE |
batch79-nova-pro-m3-r2AWS Enterprise Discount Program (EDP) drawdown | Qualifies for enterprise commitment spend | Inference bills draw down directly against existing annual AWS enterprise commitments | EDP drawdown confirmed | VERIFIED_DETERMINISTIC |
batch79-nova-pro-m3-r3Provisioned throughput capacity reservation | Bedrock model units reservation | Guarantees dedicated throughput with predictable latency for mission-critical workloads | Throughput guaranteed | VALIDATED_OBSERVED |
batch79-nova-pro-m3-r432K Output token ceiling headroom | 32,768 max completion token limit | Permits long-form report and document synthesis without truncation | Output limit confirmed | VERIFIED_DETERMINISTIC |
batch79-nova-pro-m3-r5Hybrid cascade deployment with Nova Lite & Micro | Micro triages, Lite extracts, Pro reasons | Optimizes AWS spend while maintaining end-to-end security governance | Cascade verified | MEASURED_ACTIVE |
batch79-nova-pro-m3-r6Zero egress fee within AWS cloud regions | Same-region AWS resource invocation | Zero data transfer egress fees when invoked from EC2/Lambda in same AWS region | Egress fee = $0.00 | VALIDATED_OBSERVED |
First-party provenance: Amazon Nova on Bedrock documentation; verification date 2026-09-08. Missing or conflicting joins fail closed.
What are Amazon Nova Pro's specs?
| Context window | 300K tokens |
| Max output | 33K tokens |
| Modalities | text, vision, audio |
| Extended thinking | No |
| Released | 2024-12 |
| Knowledge cutoff | 2024-10 |
| Provider | Amazon |
Verified 2026-08-14 — source.
Where does Amazon Nova Pro rank?
What are Amazon Nova Pro's strengths?
- Capable multimodal mid-tier model
- Long context for a Bedrock-native model
- Native audio and video input
What else should you know about Amazon Nova Pro?
What are common questions about Amazon Nova Pro?
What is Amazon Nova Pro's context window?
Amazon Nova Pro has a 300K-token context window and a 33K-token max output — the 23rd-largest context of the 39 current models we track. Source: https://docs.aws.amazon.com/nova/latest/userguide/what-is-nova.html, verified 2026-08-14.
Does Amazon Nova Pro support vision or audio input?
Yes — Amazon Nova Pro accepts vision and audio input in addition to text.
Does Amazon Nova Pro have a reasoning or extended-thinking mode?
No — Amazon Nova Pro does not expose a separate reasoning/extended-thinking mode.
When was Amazon Nova Pro released, and what is its knowledge cutoff?
Amazon Nova Pro was released 2024-12 with a knowledge cutoff of 2024-10.
How much does Amazon Nova Pro cost, and who provides it?
Amazon Nova Pro is served by Amazon at $1.40/M blended tokens (3:1 input:output) — the 15th-cheapest of 39 current models. Full pricing breakdown: /llm-api-pricing/nova-pro.