Nano Banana 2
kenari/nano-banana-2 · family gemini-flash · in: text, image, video, pdf → out: text, image · price provider-declared (via models.dev). Capabilities: reasoning text image video pdf.
free
Blended $/1M (3:1)
free
Input $/1M
free
Output $/1M
free
1M in + 1M out
—
Cache read $/1M
#439 of 8,002
Cheapest rank
- Provider
- Kenari (
kenari) - Kind
- text (outputs text, image)
- Context
- 131,072 tokens
- Max output
- 32,768 tokens
- Capabilities
- reasoning text image video pdf
Same model, other providers
“Nano Banana 2” is offered by 10 providers — cheapest-effective first. This one (Kenari) is the cheapest.
| Provider | Input | Output | Blended | Context |
|---|---|---|---|---|
| Kenari (this) | free | free | $0.00 | 131k |
| Requesty | $0.5 | $2 | $0.88 | 131k |
| Kilo Gateway | $0.5 | $3 | $1.13 | 131k |
| Abacus | $0.5 | $3 | $1.13 | 1M |
| OpenRouter | $0.5 | $3 | $1.13 | 131k |
| Vercel AI Gateway | $0.5 | $3 | $1.13 | 131k |
| AIHubMix | $0.5 | $3 | $1.13 | 131k |
| Eden AI | $0.5 | $3 | $1.13 | 66k |
| $0.5 | $60 | $15.38 | 66k | |
| Vertex | $0.5 | $60 | $15.38 | 131k |
Call it
jishie indexes & prices models; it does not proxy inference.
Most providers are OpenAI-compatible — point base_url at Kenari and pass this model id:
curl $BASE_URL/v1/chat/completions \
-H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" \
-d '{"model":"nano-banana-2","messages":[{"role":"user","content":"hello"}]}'
Route programmatically with MCP find_model / get_model, or fetch this record free (no wallet):