Nova Lite
A fast, very low-cost multimodal model for high-volume Bedrock workloads where Nova Pro would be overkill.
At a glance
Released
Apr 22, 2026
Provider
Amazon
Context window
300,000 tokens
Max output
16,000 tokens
Modalities
Text + image + video → text
Knowledge cutoff
February 2026
API model ID
nova-lite-v1License
Proprietary (API-only via AWS Bedrock)
Pricing
Standard
Input$0.06 / 1M
Output$0.24 / 1M
Batch (50% off)
Input$0.03 / 1M
Output$0.12 / 1M
A typical 1,500-token prompt with a 650-token answer costs ≈ $0.0002 on the standard tier. Run your own prompt through the calculator →
What's new vs the previous model
Compared withNova Lite (2024) (December 2024)
- Better instruction following and multimodal grounding
- Lower latency at scale
- Still noticeably weaker than Nova Pro on complex reasoning
Best use cases
High-volume classification
Routes, tags, and triages large request streams at minimal per-call cost.
Chat assistants
Powers responsive customer-facing assistants with image understanding.
Content summarization
Condenses long documents and transcripts quickly and cheaply.
Strengths & limitations
Strengths
- Very cheap
- Low latency
- Multimodal input
Limitations
- Limited depth on hard reasoning
- API-only
How to test it
- No-code: open the Amazon Bedrock console and run a prompt in the playground.
- API: create a key in the Amazon Bedrock console and call the API (example below).
aws bedrock-runtime converse \
--model-id nova-lite-v1 \
--messages '[{"role": "user", "content": [{"text": "Explain tokenization in one paragraph."}]}]'