Amazon: Nova Lite 1.0

amazon/nova-lite-v1 · AmazonDeclared and measured agree

Tokenizer lineage (measured): Nova
Declared tokenizer tag: Nova
Entered catalogue: 5 December 2024
Last tested: 21 August 2026
Evidence confidence: Very high

Why very high: an exact fingerprint shared with 2 models and an identical API contract inside the same group: two signals of different kinds agree.

Evidence

✓ exact tokenizer signature shared with 2 models
✓ API-surface contract identical to a group member

Compare with Amazon: Nova Micro 1.0

Report a correction

Identity statement

Amazon: Nova Lite 1.0 is Amazon's model. Its measured tokenizer sits with the Nova group, which is evidence that it builds on the Nova tokenizer. Tokenizer reuse is normal engineering, open vocabularies travel between labs, and this says nothing against Amazon's authorship of the model itself.

What we checked

Four different kinds of evidence, and what each one showed

We checked	What we found	Where it came from
How it counts tokens	Counts every one of the 50 test strings exactly like 2 other models	tk_76e424dd
How its API is set up	Set up identically to 1 other listing, field for field	api_3a1a14ff
How much it can read and write	Reads up to 300,000 tokens at once, writes up to 5,120	its own listing
Who served our requests	Amazon Bedrock	we measured it, 21 August 2026

Shares this fingerprint

Exact signature first; template-boundary shifts of the same signature beneath

Amazon: Nova Micro 1.0amazon/nova-micro-v1exact · 50/50 Amazon: Nova Pro 1.0amazon/nova-pro-v1exact · 50/50

Closest measured models

Agreement across the mutually clean test strings

Amazon: Nova Micro 1.0amazon/nova-micro-v150/50 · 100%Amazon: Nova Pro 1.0amazon/nova-pro-v150/50 · 100%Amazon: Nova Premier 1.0amazon/nova-premier-v140/50 · 80%NVIDIA: Nemotron 3.5 Content Safetynvidia/nemotron-3.5-content-safety17/50 · 34%NVIDIA: Nemotron 3.5 Content Safety (free)nvidia/nemotron-3.5-content-safety:free14/50 · 28%Poolside: Laguna S 2.1poolside/laguna-s-2.113/50 · 26%

Click a row to open the full comparison.

The measured fingerprint

What each test string cost this model, in tokens

We sent Amazon: Nova Lite 1.0 fifty short pieces of text and recorded what each one cost it in tokens. The bars below are those costs. Two models built on the same tokenizer produce the same bars; a model built on a different one produces a different set, which is what makes this a fingerprint.

All 50 rows

English and whitespace

en-prose13

en-long15

spaces-203

spaces-603

tabs-203

newlines-204

mixed-ws5

Digits

digits-910

digits-1213

digits-3031

digits-sep13

float-long23

CJK

zh-common17

zh-long24

zh-rare28

ja-kana9

ja-kanji13

ko19

Other scripts

ru14

ar10

he14

hi11

th22

el17

Emoji

emoji-basic5

emoji-skin14

emoji-zwj-family20

emoji-zwj-x358

emoji-flags24

emoji-prof30

Rare Unicode

math29

boxdraw36

combining9

cjk-ext-b17

surrogates55

zalgo34

rtl-mix8

Code

py-code28

py-indent15

json28

html16

regex48

camel6

snake11

Repetition and encodings

rare-word-x520

repeat-tok40

base6436

hex16

url23

uuid37

Overhead subtracted: 0 prompt tokens.

Declared record

What the catalogue claims about this model

Amazon Nova Lite 1.0 is a very low-cost multimodal model from Amazon that focused on fast processing of image, video, and text inputs to generate text output. Amazon Nova Lite...

Modalities: text+image->text
Declared tokenizer: Nova
Prompt price: $0.06 / M tokens
Completion price: $0.24 / M tokens

Supported parameters

max_tokensstoptemperaturetoolstop_ktop_p

defaults: {}

Catalogue entry

History

Every observation, kept as taken

5 December 2024Enters the OpenRouter catalogue declared as Nova.
21 August 2026Fingerprinted in the YFarmX catalogue sweep · 50 of 50 strings measured clean.
21 August 2026YFarmX assessment: consistent with the Nova family, very high confidence.