Amazon Nova Premier
DeprecatedMost capable Nova 1 model: complex reasoning, agentic workflows, model distillation. Reasoning supported. 1M-token context, 25K max output. Model card states lifecycle = 'Legacy' and 'Model EOL date: September 14, 2026'; the row is held at 'deprecated' (= Legacy), which is what AWS declares. Until 2026-09-15 this note read that EOL date as a firm retirement — that reading did not survive the date. Re-read first-hand on 2026-09-15, the day AFTER the stated EOL: the card still says 'Model lifecycle: Legacy' and still carries the same EOL date, so the status stays 'deprecated'. The signal used is the card's own lifecycle field; a shutdown date that has merely passed is not a retirement signal on its own. The same card also states 'EOL no sooner than: Oct 31, 2026' beside 'Model EOL date: September 14, 2026' — two first-party fields that contradict each other. Both are recorded verbatim here and neither is resolved into our typed fields. Corroboration and its limit: the public Bedrock Price List API still carries every Nova Premier rate in us-east-1, but that price book was published 2026-09-11, BEFORE the EOL date, so it evidences 'still sold on 09-11', not 'still sold today'. NOTE: model card shows 'Model launch date: Oct 31, 2025' which conflicts with the public GA of Nova Premier (Apr/May 2025); the Oct 31 2025 date may reflect a card/version update rather than original GA — flagged as uncertain. Only us. geo inference ID (us.amazon.nova-premier-v1:0). The Price List API (us-east-1, 2026-09-11 book) publishes the rates per 1K tokens as $0.0025/1K input, $0.0125/1K output and $0.000625/1K cache read, which is exactly the $2.50 / $12.50 / $0.625 per 1M this row carries. That last sentence had been cut mid-word, immediately after the first per-1K figure, since the repo's initial import, which truncated this note at exactly 600 characters; the tail is restored from the Price List API, not reconstructed.
Amazon Nova Premier by Amazon costs $2.50 per 1M input tokens and $12.50 per 1M output tokens ($5/1M blended), with a 1M (1.000.000-token) context window. It is deprecated, shutting down 14 Sep 2026; the recommended replacement is Amazon Nova 2 Lite.
Last verified: 26 Sep 2026 · sourced from official provider documentation
Retired on 14 Sep 2026
◎ You're on the watch list. We'll ping you the moment a model launches, changes price, or gets deprecated.
✕ Couldn't subscribe right now — please try again in a moment.
Free forever · no spam · unsubscribe anytime.
This model is deprecated. Recommended migration: Amazon Nova 2 Lite.
Read the Amazon Nova Premier migration guide → replacement & cheapest alternatives
Source: Amazon official documentation ↗
Amazon Nova Premier — questions & answers
How much does Amazon Nova Premier cost?
Amazon Nova Premier is priced at $2.50 per 1M input tokens and $12.50 per 1M output tokens ($5/1M blended at a 3:1 input-to-output ratio), with cached input at $0.625 per 1M tokens.
What is the context window of Amazon Nova Premier?
Amazon Nova Premier has a 1.000.000-token context window (1M), with up to 25.000 output tokens per request.
Is Amazon Nova Premier being deprecated or retired?
Amazon Nova Premier is deprecated. Its scheduled shutdown date is 14 Sep 2026. The recommended migration is Amazon Nova 2 Lite.
Does Amazon Nova Premier support image or vision input?
Yes — Amazon Nova Premier accepts image input. Listed modalities: text, image, video.
Can I self-host Amazon Nova Premier?
No — Amazon Nova Premier is proprietary. Its weights are not publicly released, so it is available only through Amazon's API (or hosted partners), not for self-hosting.
What is the API model string for Amazon Nova Premier?
The API model identifier for Amazon Nova Premier is "amazon.nova-premier-v1:0" when calling Amazon's API.
What is Amazon Nova Premier's knowledge cutoff?
Amazon Nova Premier's training knowledge cutoff is Oct 2024.
Track Amazon Nova Premier price & status changes
New models, price cuts, and deprecations — a short email when something actually changes. No spam, unsubscribe anytime.
The last few changes we caught — this is what lands in your inbox:
- DeprecatingGoogle re-pointed the Gemini 2.5 TTS migration path past the 3.1 preview and onto the new GA 3.8 TTS models
- New modelGemini 3.8 Flash TTS is GA - a flagship text-to-speech model at under half the audio rate of the preview it replaces, on a two-column price schedule
- New modelGemini 3.8 Flash-Lite TTS is GA - Google's cost-efficient TTS, which Google says was built to replace Gemini 3.1 Flash TTS
◎ You're on the watch list. We'll ping you the moment a model launches, changes price, or gets deprecated.
✕ Couldn't subscribe right now — please try again in a moment.
Free forever · powered by the same data on this page.