Gemini 2.5 Flash Lite Max Thinking (2025-09-25)
Parameters
-
Context Length
1,048.576K
Modality
Multimodal
Architecture
Dense
License
Proprietary
Release Date
25 Sept 2025
Knowledge Cutoff
-
Technical Specifications
Attention Structure
Multi-Head Attention
Hidden Dimension Size
-
Number of Layers
-
Attention Heads
-
Key-Value Heads
-
Activation Function
-
Normalization
-
Position Embedding
Absolute Position Embedding
Gemini 2.5 Flash Lite Max Thinking (2025-09-25)
Lightweight Gemini 2.5 with max thinking mode optimized for cost-efficiency. Decent coding (65.39 LiveBench Coding) and mathematics (64.90) performance at reduced computational cost. September 2025 version. Ideal for high-volume applications requiring thinking transparency without premium pricing.
About Gemini 2.5
Google's advanced multimodal models with native understanding of text, images, audio, and video. Features massive context windows up to 2.1M tokens, max thinking modes for complex reasoning, and optimized variants for different performance/cost tradeoffs. Includes Pro, Flash, and Flash Lite variants with configurable thinking capabilities for transparent reasoning.
Other Gemini 2.5 Models
Evaluation Benchmarks
Rank
#70
| Benchmark | Score | Rank |
|---|---|---|
Data Analysis LiveBench Data Analysis | 0.68 | 28 |
Mathematics LiveBench Mathematics | 0.65 | 37 |
Coding LiveBench Coding | 0.65 | 42 |
Reasoning LiveBench Reasoning | 0.36 | 53 |
Agentic Coding LiveBench Agentic | 0.02 | 53 |
Rankings
Overall Rank
#70
Coding Rank
#45