Model hub
Browse available model capabilities behind the unified gateway and quickly choose the best fit for your product by scenario.
This page highlights the Gemini image models currently prioritized by BananaRouter, including gemini-3.1-flash-image-preview, gemini-3-pro-image-preview, and gemini-2.5-flash-image.
gpt-image-2
From ¥0.04/call
gpt-image-2
gpt-5.4
Input ¥1.75/1M tokensOutput ¥10.50/1M tokensCache read ¥0.725/1M tokens
gpt-5.4
gpt-5.5
Input ¥3.50/1M tokensOutput ¥21.00/1M tokensCache read ¥0.225/1M tokens
gpt-5.5
gpt-5.4-mini
Input ¥0.53/1M tokensOutput ¥3.15/1M tokensCache read ¥0.20/1M tokens
gpt-5.4-mini
gpt-image-1.5
From ¥0.04/call
gpt-image-1.5
doubao-seedance-2.0
From ¥26.60/1M tokens
480p, 720p, 1080p
GLM-5.2
Input ¥8.00/1M tokensOutput ¥28.00/1M tokens
GLM-5.2
kimi-k2.6
Input ¥6.50/1M tokensOutput ¥27.00/1M tokens
kimi-k2.6
kimi-k2.7-code
Input ¥6.50/1M tokensOutput ¥27.00/1M tokens
kimi-k2.7-code
qwen3.7-plus
Input ¥6.00/1M tokensOutput ¥24.00/1M tokens
qwen3.7-plus
qwen3.7-max
Input ¥12.00/1M tokensOutput ¥36.00/1M tokens
qwen3.7-max
qwen3.6-plus
Input ¥8.00/1M tokensOutput ¥48.00/1M tokens
qwen3.6-plus
qwen3.6-flash
Input ¥4.80/1M tokensOutput ¥28.80/1M tokens
qwen3.6-flash
MiniMax-M2.5
Input ¥2.10/1M tokensOutput ¥8.40/1M tokens
MiniMax-M2.5
gemini-3.1-flash-lite-image
From ¥0.07/call
1K
Grok
From ¥0.00/call
480p、720p
seedance-2.0
From ¥0.00/call
720p
gpt-5.6-Sol
Input ¥4.20/1M tokensOutput ¥19.00/1M tokensCache read ¥0.30/1M tokensCache write ¥3.00/1M tokens
gpt-5.6-Sol
gpt-5.6-Terra
Input ¥2.10/1M tokensOutput ¥9.60/1M tokensCache read ¥0.15/1M tokensCache write ¥1.35/1M tokens
gpt-5.6-Terra
gpt-5.6-Luna
Input ¥0.84/1M tokensOutput ¥3.78/1M tokensCache read ¥0.06/1M tokensCache write ¥0.60/1M tokens
gpt-5.6-Luna
claude-fable-5
Input ¥26.00/1M tokensOutput ¥129.00/1M tokensCache read ¥3.00/1M tokensCache write (5m) ¥30.00/1M tokensCache write (1h) ¥48.00/1M tokens
claude-fable-5
sora-2
From ¥1.00/call
720x1280, 1280x720
gemini-3.1-flash-image
From ¥0.10/call
Fast Gemini image generation preview model with multi-resolution output, suitable for creative iteration and UI drafts.
gemini-3-pro-image
From ¥0.22/call
Higher-quality Gemini image model with stronger composition and detail for production-oriented visuals.
gemini-2.5-flash-image
From ¥0.07/call
Balanced Gemini 2.5 image model for everyday generation workloads with reliable latency.
gemini-3.1-pro-preview
Input ¥3.50/1M tokensOutput ¥21.00/1M tokensCache read ¥0.40/1M tokensInput (>200k) ¥5.00/1M tokensOutput (>200k) ¥36.00/1M tokensCache read (>200k) ¥0.80/1M tokens
gemini-3.1-pro-preview
gemini-3-flash-preview
Input ¥0.88/1M tokensOutput ¥5.25/1M tokens
gemini-3-flash-preview
gemini-3.1-flash-lite
Input ¥0.44/1M tokensOutput ¥2.63/1M tokens
gemini-3.1-flash-lite
gemini-3.5-flash
Input ¥2.63/1M tokensOutput ¥15.75/1M tokens
gemini-3.5-flash
gemini-2.5-flash
Input ¥0.53/1M tokensOutput ¥4.38/1M tokens
gemini-2.5-flash
gemini-2.5-pro
Input ¥2.19/1M tokensOutput ¥17.50/1M tokens
gemini-2.5-pro
gemini-2.5-flash-lite
Input ¥0.18/1M tokensOutput ¥0.70/1M tokens
gemini-2.5-flash-lite
claude-opus-4-8
Input ¥7.00/1M tokensOutput ¥35.00/1M tokensCache read ¥0.60/1M tokensCache write (5m) ¥6.00/1M tokensCache write (1h) ¥9.60/1M tokens
claude-opus-4-8
claude-sonnet-5
Input ¥8.00/1M tokensOutput ¥40.00/1M tokensCache read ¥0.50/1M tokensCache write (5m) ¥7.00/1M tokensCache write (1h) ¥11.20/1M tokens
claude-sonnet-5
claude-opus-4-7
Input ¥7.00/1M tokensOutput ¥35.00/1M tokensCache read ¥0.60/1M tokensCache write (5m) ¥6.00/1M tokensCache write (1h) ¥9.60/1M tokens
claude-opus-4-7
claude-opus-4-6
Input ¥7.00/1M tokensOutput ¥35.00/1M tokensCache read ¥0.60/1M tokensCache write (5m) ¥6.00/1M tokensCache write (1h) ¥9.60/1M tokens
claude-opus-4-6
claude-sonnet-4-6
Input ¥4.20/1M tokensOutput ¥21.00/1M tokensCache read ¥0.50/1M tokensCache write (5m) ¥7.00/1M tokensCache write (1h) ¥11.20/1M tokens
claude-sonnet-4-6
deepseek-v4-flash
Input ¥1.10/1M tokensOutput ¥2.20/1M tokens
deepseek-v4-flash
deepseek-v4-pro
Input ¥3.30/1M tokensOutput ¥6.60/1M tokens
deepseek-v4-pro