Gemini 3.5 Flash-Lite
Gemini 3.5 Flash-Lite is Google's generally available, high-throughput Gemini model for latency-sensitive and repeated workloads. Its Gemini API model ID is gemini-3.5-flash-lite.
Flash-Lite gives teams a distinct routing tier for tasks that need multimodal understanding, tools, and long context but cannot justify the latency or output price of the main Flash model on every call.
Google announced Gemini 3.5 Flash-Lite on July 21, 2026 and documents it as GA. It supports text, image, audio, and video input, text output, a 1M-token context window, 64K maximum output, thinking, and built-in tools including Computer Use. Google lists a minimal default thinking level, standard pricing of $0.30 per million input tokens and $2.50 per million output tokens, and vendor-measured throughput of 350 output tokens per second on the Artificial Analysis Index.
- Extract and classify structured data at high volume.
- Process large document, translation, search, and summarization queues.
- Run repeated subagent tasks under a stronger orchestrator model.
- Serve low-latency multimodal workflows with configurable thinking effort.
Gemini 3.5 Flash-Lite targets agentic search, document processing, extraction, classification, translation, structured JSON parsing, and autonomous subagent execution. Google recommends minimal thinking for maximum throughput and medium or high thinking for multi-step tool use.
The GA model ID is gemini-3.5-flash-lite. Google documents a 1M-token context window, 64K maximum output, multimodal input, text output, and standard pricing at $0.30 per million input tokens and $2.50 per million output tokens.
Choose 3.5 Flash-Lite when throughput, latency, and per-call cost dominate. Choose 3.6 Flash when difficult coding, spatial or multimodal reasoning, and multi-step orchestration need stronger quality. A production router should compare task success and retry rate, not route only by token price.
Gemini 3.5 Flash-Lite FAQ
Common questions about Gemini 3.5 Flash-Lite.
What is the Gemini 3.5 Flash-Lite API model ID?+
The generally available Gemini API model ID is gemini-3.5-flash-lite.
How much does Gemini 3.5 Flash-Lite cost?+
Google lists standard pricing at $0.30 per million input tokens and $2.50 per million output tokens. Check the current pricing page for batch, caching, or provider-specific rates before estimating production cost.
When should I use Gemini 3.5 Flash-Lite instead of 3.6 Flash?+
Use 3.5 Flash-Lite for high-volume extraction, classification, document processing, search, and repeated subagent work where latency and cost are primary. Use 3.6 Flash when the task needs stronger coding, multimodal reasoning, or orchestration quality.