Chuyển tới nội dung chính

🧠 Gemini Model Guide

Orchable supports a wide range of Gemini models, each optimized for different stages of the orchestration pipeline.


🚀 Model Comparison

ModelPrimary Use CaseContext WindowKey Features
Gemini 3 ProHighest reasoning, complex coding1M+ tokensState-of-the-art reasoning, SOTA vibe-coding
Gemini 3 FlashBalanced speed & intelligence1M+ tokensPro-grade reasoning at Flash speed
Gemini 2.5 ProDeep STEM, math, long context1M+ tokensOptimized for analytical depth
Gemini 2.5 FlashProduction workhorse1M+ tokensBest price-performance; ultra-reliable
Gemini 2.5 Flash LiteHigh-volume micro-tasks1M+ tokensFastest, most affordable
Gemini 2.0 FlashLegacy support / Stable Gen1M tokensMature, stable performance

🛠️ Capability Matrix

CapabilityGemini 3 SeriesGemini 2.5 SeriesGemini 2.0 Series
Structured Output (JSON)
Thinking / CoT✅ (Multi-level)✅ (Budget)✅ (Exp)
Context Caching
Code Execution
Multimodal (Audio/Video)
Live API (Streaming)

📍 Choosing a Model

  • Planning Stage: Use Gemini 3 Flash or Gemini 3 Pro for high-quality orchestrations.
  • Processing Stage: Use Gemini 2.5 Flash for reliable, consistent output.
  • Filtering Stage: Use Gemini 2.5 Flash Lite to save cost and time.
  • RAG Stage: Use Gemini Embedding 1 for all vector retrieval tasks.

Last Updated: 2026-02-24