Gemini 3.5 Flash-Lite
Google · Multimodal · Released Jul 2026
Google's lightweight text model designed for efficient subagent execution within multi-agent workflows, with a 1M token context window.
- Strengths
- Fast inference and high throughput make it well-suited for handling focused, repetitive tasks in agentic systems without the overhead of larger models.
- Best for
- Subagents that execute specialized tasks within complex multi-agent workflows where speed and efficiency take priority over raw capability.
- Limitations
- As a lightweight model, it trades general reasoning depth for speed, making it less suitable for tasks requiring sophisticated analysis or broad knowledge.
Input / 1M
$0.3
Output / 1M
$2.50
Cached input / 1M
$0.03
Context window
1.05M
Also billed
- Cache write
- $0.0833 / 1M
- Audio input
- $0.3 / 1M
Price history
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 21 Jul 2026 | $0.3 | $2.50 | $0.03 | Imported from OpenRouter | openrouter.ai |