model releaseGoogle DeepMind

Google DeepMind releases Nano Banana 2 Lite at $0.034 per 1K image with 4-second generation, opens Gemini Omni Flash API

TL;DR

Google DeepMind released Nano Banana 2 Lite (gemini-3.1-flash-lite-image), its fastest image generation model with 4-second text-to-image latency priced at $0.034 per 1K-resolution image. The company also opened developer access to Gemini Omni Flash (gemini-omni-flash-preview) for video generation and editing at $0.10 per second of output.

3 min read
0

Google DeepMind releases Nano Banana 2 Lite at $0.034 per 1K image with 4-second generation, opens Gemini Omni Flash API at $0.10 per video second

Google DeepMind released Nano Banana 2 Lite (gemini-3.1-flash-lite-image), its fastest image generation model with 4-second text-to-image latency priced at $0.034 per 1K-resolution image. The company also opened developer access to Gemini Omni Flash (gemini-omni-flash-preview) for video generation and editing at $0.10 per second of output.

Nano Banana 2 Lite targets high-throughput developer workflows where speed and cost matter more than maximum quality. According to Google, it's positioned as the recommended replacement for the original Nano Banana (gemini-2.5-flash-image).

Nano Banana 2 Lite specifications

  • Latency: 4 seconds for text-to-image generation
  • Pricing: $0.034 per 1K-resolution image
  • API name: gemini-3.1-flash-lite-image
  • Availability: Google AI Studio, Gemini API, Gemini Enterprise Agent Platform

Google claims the model maintains "reliable prompt adherence, strong character consistency and legible in-image text rendering" despite optimizations for speed. The company published internal benchmarks comparing Elo quality scores, latency, and cost against competitor models, though independent verification is not yet available.

The Nano Banana family now includes four models:

  • Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image): Speed-optimized
  • Nano Banana 2 (Gemini 3.1 Flash Image): Balanced performance
  • Nano Banana Pro (Gemini 3 Pro Image): Quality-optimized for professional use
  • Nano Banana (Gemini 2.5 Flash Image): Legacy model, upgrade recommended

Gemini Omni Flash video capabilities

Gemini Omni Flash, first introduced at Google I/O, is now accessible via API in public preview. The model handles video generation and editing from text, image, and video inputs with conversational refinement.

Key specifications:

  • Pricing: $0.10 per second of video output (matching Veo 3.1 Fast)
  • API name: gemini-omni-flash-preview
  • Current output length: 10 seconds (longer durations coming)
  • Availability: Google AI Studio, Gemini API

Current limitations:

  • Video references up to 3 seconds accepted but not correctly processed
  • Audio reference uploads not yet supported
  • Character consistency issues across scene changes
  • Scene extension not available

The model uses Gemini's multimodal reasoning to synchronize text, graphics, and actions in generated video. Google positions it for conversational editing workflows where users iteratively refine outputs through natural language.

Combined workflow capabilities

Google published three demo applications showing Nano Banana 2 Lite generating initial images that Gemini Omni Flash then animates:

  • Anywhere: Generates landmark backgrounds, animates into video clips
  • Space Lift: Interior design visualization with cinematic walkthroughs
  • Omni Product Studio: E-commerce product video generation

The Interactions API supports up to three sequential edits with maintained session context. Both models include SynthID watermarking for content provenance.

What this means

Google is aggressively pricing Nano Banana 2 Lite to compete in the high-volume image generation market, undercutting alternatives where speed matters more than quality. The $0.034 per 1K image pricing and 4-second latency target prototyping and draft-heavy workflows. However, without independent benchmarks, developers need to validate quality claims against their specific use cases.

The Gemini Omni Flash API opening is more significant. At $0.10 per second, it's the first major video generation model with conversational editing officially available via API, though the 10-second limit and processing bugs (especially around video references) suggest it's genuinely preview-stage. The combined image-to-video pipeline could enable new product categories in e-commerce, marketing, and content creation—if the quality and reliability hold up at scale.

Related Articles

model release

Gemini 4 Neon scores 61 on Artificial Analysis Intelligence Index, 3rd overall, at $3/$15 per 1M tokens

Google DeepMind has released Gemini 4 Neon, its fourth model in seven weeks. Artificial Analysis measures it at 61 on the Intelligence Index (xhigh), 3rd overall, with token pricing of $3/$15 per 1M input/output tokens and a 1M-token context window.

model release

OpenAI releases GPT-Live-1.1 in the API, with standby mode billing idle WebRTC time at $0.02 per minute

OpenAI has released GPT-Live-1.1 in the API. The full-duplex voice model improves multilingual accents and transcription, and a new standby mode bills idle WebRTC time at $0.02 per minute instead of $0.05.

model release

Baseten releases open-weight bison-1 and groundhog-1 decision models; bison-1 scores 76 on AA Decision Index v2.1

Baseten released bison-1 and groundhog-1, two open-weight decision models built on Gemma 4. Bison-1 scores 76 on the AA Decision Index v2.1, and both models support 128k context and image input.

model release

Thinking Machines Lab releases Inkling 2 Mini: 26B-parameter looped MoE with 3B active and 1M-token context

Thinking Machines Lab has released Inkling 2 Mini, a multimodal Mixture-of-Experts model with about 26B physical total parameters and 3B active per token. It repeats its middle layers three times, supports a 1M-token context, and ships with a DSpark drafter. Hugging Face reports it leads 32 of 48 benchmarks against four comparable models.

Comments

Loading...