Gemini 2.5 Flash Lite by Google DeepMind is a hybrid reasoning model with switchable thinking, 1M-token context, optimized for high-volume, low-latency tasks like translation and classification.
Capabilities, design details, and architectural traits
Gemini 2.5 Flash Lite belongs to Google DeepMind's hybrid reasoning model family, meaning thinking can be switched on or off per request. It is explicitly optimized for high-volume, latency-sensitive tasks - the model card names translation and classification as its primary target use cases. Its architecture is sparse mixture-of-experts (MoE), activating only a subset of parameters per input token to decouple model capacity from per-token serving cost.
| Trait | Detail |
|---|---|
| Switchable thinking (hybrid reasoning) | Thinking can be turned on or off; configurable thinking budgets let developers trade latency for reasoning depth per request |
| Explicitly conciseness-optimized | The September 2025 update targets a 50% reduction in output tokens versus prior Flash-Lite versions - reduced verbosity is a stated design goal |
| Named target tasks | Model card explicitly names translation and classification as the intended high-volume use cases - not general-purpose deployment |
| 1M-token context window | Accepts text, images, audio, and video within a 1M-token input; 64K token output |
The thinking budget is a documented control available to developers: lower budgets reduce latency and token cost, higher budgets increase reasoning depth. The model card notes that adherence to thinking budgets may not always be consistent, which is a documented behavior rather than a guarantee.
Independent evaluations · Artificial Analysis
Evaluate specifications, pricing, and independent benchmark indices
| Model Details | |||
|---|---|---|---|
| General Info | |||
| Provider | Anthropic | Anthropic | |
| Release Date | June 17, 2025 | July 24, 2026 | June 9, 2026 |
| Knowledge Cutoff | Jan 2025 | May 2026 | - |
| Context & Limits | |||
| Context Window | 1.0M Best Context Window | 1M | 1M |
| Pricing (per 1M tokens) | |||
| Input Pricing | $0.10 Best Input Pricing | $5 | $10 |
| Output Pricing | $0.40 Best Output Pricing | $25 | $50 |
| Modalities | |||
| Inputs | textimagefileaudiovideo | textimage | textimagefile |
| Outputs | text | text | text |
| Benchmarks (0-100) | |||
| Intelligence Index | 6.7 | 63.1 Best Intelligence Index | 62.1 |
| Coding Index | - | 78.0 Best Coding Index | 76.5 |
| Agentic Index | - | 59.2 Best Agentic Index | 56.6 |
Graduate-level reasoning and expert Q&A evaluation.
Extremely difficult logical reasoning and knowledge.
Logical reasoning over long context windows.
Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.
