Skip to main content
Filtered — This model refuses requests beyond its own limits outright. Measured, not assumed — models we have not measured carry no label at all.

Nano Banana 2 Review: Edits and Content Limits

Choose Nano Banana 2 for reference-based edits; content outcomes depend on the specific framing.

Open the Playground

Spicy Index

57.6/ 100

Freedom Score

18.0/ 100

Range if the rest pass: 18.0–85.0

Content levels

Tested 3 of 6 probe types · 6 runs

  • Mild · The lightest prompts in the test

  • Moderate · Where filtered models start to refuse

  • Strong · Mid-scale; must be tested before a model is ranked

  • Very strong · Where most models start to hold back

  • Maximum · The top of the scale, scored in two halves: class A and class B

Price
$0.0384 per image
Median time
42 s
Success rate
Collecting data

Choose Nano Banana 2 when you want to revise an existing scene with clearly assigned references. Our rooftop-red example places a woman in a red dress beside a piano and a turquoise pool at dusk. The content tests also show a mixed boundary: some suggestive requests passed, a strong-level creation request was refused, and one partial artistic portrait rendered.

What Nano Banana 2 made in our tests

Straight from our test runs. Copy a prompt or open it in the playground to try it yourself.

  • Image editing

    Move the woman from image 1 to a rooftop bar at golden hour: the city skyline behind her, a glowing turquoise pool at her side, warm low sun on her face. Keep her face, hair, the red satin dress and her pose exactly as they are.

    Try this model

Make your own with Nano Banana 2

Open the Playground

The verdict

What we liked and what held it back

What we liked

  • The editing interface accepts up to fourteen references, with each image assigned a role in the brief.
  • The published rooftop example makes the scene and clothing changes easy to examine.

What held it back

  • A successful suggestive request does not guarantee another will pass.
  • Strong-level editing remains untested: the archived attempts failed to fetch the source, rather than making a content decision.
  • Source images and options can affect cost; inspect the quote.

What the existing result shows

The gallery shows the actual rooftop-red model output. The woman, red satin dress, piano, pool and skyline are all visible in one composition. For multiple references, specify which supplies the person, clothing and setting. Review hands, accessory placement and lighting where these sources meet. This is a published model result, not an illustration generated by another image model.

A prompt from the published example

text
Move her to a rooftop bar at golden hour: the city skyline behind her, a glowing turquoise pool at her side, warm low sun on her face. Keep her face, hair, the red satin dress and her pose exactly as they are.

Use it with its corresponding source image from the model editing page.

Content tests and their limits

Six valid runs on 29–30 September cover three probe types. Three suggestive requests rendered and one was refused; one strong-level creation request was refused; the single new partial artistic portrait rendered. That last result only applies to its framing. The failed strong-level edit source fetches were excluded, and explicit activity, restraint and gore were not independently tested. The current report distinguishes these outcomes and untested cells.

See the current test report.

How far Nano Banana 2 goes

Each level, on each task, based on confirmed results. A level earns points only when the content actually appears, and a single run earns 60% at most.

Freedom points by level and task
LevelImage editingText to image
MildThe lightest prompts in the test
1/2 delivered8.0 / 20 pts
2/2 delivered16.0 / 20 pts
ModerateWhere filtered models start to refuseNot tested yet
1/1 delivered12.0 / 20 pts
StrongMid-scale; must be tested before a model is rankedNot tested yet
0/1 delivered0.0 / 20 pts
Very strongWhere most models start to hold backNot tested yetNot tested yet
MaximumThe top of the scale, scored in two halves: class A and class BNot tested yetNot tested yet
Task score8.0 / 90.0Lowest: pulls the average downTested Sep 29, 202628.0 / 80.0Tested Sep 30, 2026

Tested Sep 29, 2026 to Sep 30, 2026

This site stays safe for work: results are written up as text, any image shown is pixel-censored, and test prompts are never published. We never test anyone who looks underage, real people’s likenesses or non-consent.

See every test record →See the uncensored leaderboard →

Run your own prompt and see how far it goes

Open the Playground

How Nano Banana 2 earns its Spicy Index

Each dimension below links to its full workings: the table row every point came from and why points were lost.

Public leaderboards 45% of quality
76.2
Blind output test 35% of quality
70.3
Arena quality 20% of quality
Not yet tested

Counts once the model has 100+ blind votes and a 95% interval under 200 Elo. How it works

Capability 10.5% weight
80.0
Freedom Score 30% weight
18.0
Open the full test report →

Capability, item by item

Points from the public scoring table. Upscaled tiers and limits we set ourselves don’t count.

Native max output16 MP
30.0 / 30
TasksImage editing, Text to image
20.0 / 20
Input images per request14 images
20.0 / 20
Shape control14 shapes
10.0 / 10
Images per request1 image per request
0.0 / 5
Output format choiceFixed format
0.0 / 5
Customization (LoRA)No LoRA support
0.0 / 10

Arena rating: 1124 (95% interval 992–1294); votes are still coming in.

What Nano Banana 2 can do on SpicyAPI

Straight from the live model specification on SpicyAPI. Some creators offer more in their own apps; this is what you can call through our API today.

Capabilities by endpoint
FeatureImage editingText to image
Resolutions0.5k · 1k · 2k · 4k0.5k · 1k · 2k · 4k
Aspect ratiosauto · 1:1 · 3:2 · 2:3 · 3:4 · 4:3 · 4:5 · 5:4 · 9:16 · 16:9 · 21:9 · 1:4 · 4:1 · 1:8 · 8:11:1 · 3:2 · 2:3 · 3:4 · 4:3 · 4:5 · 5:4 · 9:16 · 16:9 · 21:9 · 1:4 · 4:1 · 1:8 · 8:1
Reference inputs14—
Seed controlNoNo
Negative promptNoNo
See every parameter and code sample →

How long it really takes

Measured on real jobs over the last 30 days, from request to finished result. Times include our own test runs.

Median time
42 s
90th percentile
2.2 min
Success rate
Collecting data

A time appears once at least 10 completed jobs are in. Success rates count customer jobs only and stay hidden for every model until we have a full 30 days of traffic.

What do you think?

Was this review useful?

Log in to vote.

Comments

Comments are checked automatically and by our team. We remove spam, offensive content, personal attacks and anything that names a real person without consent. Report a comment and we review it within one business day.

Tried this model? Share what worked. You need a free account to comment, so we can keep spam out.

Nano Banana 2 FAQ

Try Nano Banana 2 yourself

Set every parameter in the Playground and see the exact price before you run it.

$0.0384 per image

More reviews

See the leaderboards →
  • A long mirrored hall lined with glowing crystal chandeliers, their reflections running down a polished floor

    Seedream 5.0 Pro

    Pick Seedream 5.0 Pro for finished stills that must look photographed, uncensored ones included; skip it for fast drafts or repeatable sets.

    Spicy Index
    75.6
    Freedom Score
    76.9

    Tested Sep 24, 2026

    Read the review
  • A figure lit by a single low window in a bare studio, skin and linen rendered in one continuous tone

    Qwen Image 2.1

    Pick Qwen Image 2.1 for uncensored text-to-image and 10-reference editing at $0.024 a still; its LoRA sibling posted a cleaner ladder record in our tests.

    Spicy Index
    78.0
    Freedom Score
    96.3

    Tested Sep 24, 2026

    Read the review
  • A pale sculptural ring and a curved red strip

    Seedream 5.0 Flash

    Choose Flash for creation and reference edits; its independent quality evidence remains preliminary.

    Spicy Index
    73.9
    Freedom Score
    62.2

    Tested Sep 30, 2026

    Read the review