Activefastmultimodal Proprietary

Gemini 2.0 Flash

by Google· Released December 2024· Cutoff August 2024

Gemini 2.0 Flash is Google's next-generation multimodal model offering low latency and enhanced performance across text, image, audio, and video inputs. It introduces native tool use, including Google Search grounding and code execution, making it ideal for agentic and real-time applications.

Official Site API Docs

Input cost

$0.10 per 1M tokens

Output cost

$0.40 per 1M tokens

Context window

1M tokens

Max output

8192 tokens

Modalities

textimageaudiovideo

License

proprietary

Capabilities

Multimodal understanding (text, image, audio, video)Function CallingCode GenerationStreamingJSON ModeGrounding with Google SearchCode ExecutionSystem Instructions

Best For

High-speed, cost-effective multimodal tasks and agentic workflows requiring low latency.

Strengths

Very low latency and high throughput
1M token context window
Native multimodal support (text, image, audio, video)
Built-in tool use (search, code execution)
Competitive pricing

Limitations

Not as powerful as Gemini 2.0 Pro for complex reasoning
Limited to 8192 output tokens
May not support all languages equally
Newer model with evolving ecosystem

Use Cases

Real-time chatbots and virtual assistants

Multimodal content analysis (images, videos, audio)

Code generation and debugging

Data extraction and structured output

Agentic applications with tool use

Content summarization and translation

Interactive learning and tutoring

Improvements Over Previous Model

Lower latency and higher throughput compared to Gemini 1.5 Flash
Context window increased from 1M tokens (1.5 Flash) to 1M tokens (same, but with improved retrieval)
Native multimodal input (image, audio, video) without separate preprocessing
Built-in tool use (Google Search grounding, code execution) for agentic capabilities
Pricing reduced: input $0.10/M tokens vs $0.075/M tokens (1.5 Flash) but output $0.40/M tokens vs $0.30/M tokens (1.5 Flash) — slightly higher output cost but more capabilities
Improved coding and reasoning benchmarks over 1.5 Flash
Supports JSON mode and system instructions natively

Back to all models

Improvements Over Previous Model

Lower latency and higher throughput compared to Gemini 1.5 Flash

Context window increased from 1M tokens (1.5 Flash) to 1M tokens (same, but with improved retrieval)

Native multimodal input (image, audio, video) without separate preprocessing

Built-in tool use (Google Search grounding, code execution) for agentic capabilities

Pricing reduced: input $0.10/M tokens vs $0.075/M tokens (1.5 Flash) but output $0.40/M tokens vs $0.30/M tokens (1.5 Flash) — slightly higher output cost but more capabilities

Improved coding and reasoning benchmarks over 1.5 Flash

Supports JSON mode and system instructions natively