Log in
← All models

Gemini 3.1 Flash Lite

Budget

One of the cheapest 1M-context models — built for cheap, high-volume calls.

Google
Maker
77/100
Overall
70
Value score
$0.63
Blended /1M
1.0M
Context
Proprietary
Weights
Pricing: $0.25 in · $1.50 out / 1M tokens · synced from OpenRouter · updated 2026-06-22

Scorecard

Output Quality7.4/10 · 70% community
Agentic Ability7.2/10 · 69% community
Speed9.6/10 · 80% community
Value for $8.6/10 · 75% community
Reliability8.5/10 · 75% community

Gemini 3.1 Flash Lite is Google's budget model: extremely cheap, fast, and still carrying a 1M-token context. Ideal for the high-volume, low-complexity work in an agent loop where a flagship would be wasteful.

Strengths
  • Very cheap
  • 1M context at budget price
  • Fast
Trade-offs
  • Lowest quality of the Gemini line
  • Struggles on complex multi-step tasks

Agents that run Flash Lite well

Cheap, high-volume steps that still need long context.

Best model for Gemini CLIBest model for ContinueBest model for OpenClawBest model for Hermes

Gemini 3.1 Flash Lite FAQ

Is Gemini 3.1 Flash Lite good for AI agents?
Cheap, high-volume steps that still need long context. On our editorial scoring it rates 7.2/10 for agentic ability and 7.4/10 for output quality. Very cheap.
How much does Gemini 3.1 Flash Lite cost?
Via OpenRouter, Gemini 3.1 Flash Lite is priced at $0.25 in · $1.50 out / 1M tokens — a blended rate of about $0.63 per 1M tokens for typical input-heavy agent use.
What is Gemini 3.1 Flash Lite's context window?
Gemini 3.1 Flash Lite has a 1.0M-token context window — large enough to hold sizeable codebases or document sets in a single run. Google is the maker.
Can I self-host Gemini 3.1 Flash Lite?
No — Gemini 3.1 Flash Lite is a proprietary, hosted-only model from Google. You access it through an API (e.g. OpenRouter) rather than running the weights yourself.
What are the downsides of Gemini 3.1 Flash Lite?
The main trade-offs: lowest quality of the gemini line; struggles on complex multi-step tasks. It's strongest for cheap, high-volume steps that still need long context.

Flash Lite compared

Flash Lite vs Haiku 4.5Flash Lite vs DeepSeek V4 FlashFlash Lite vs GPT-5.4 MiniFlash Lite vs Qwen3 235BFlash Lite vs Llama 4 MaverickFlash Lite vs CodestralFlash Lite vs Gemini 3.1 ProFlash Lite vs Gemini 3.5 Flash

The model is half the story — the agent is the other half

The model picks the moves; the agent runs the loop, the tools, and the guardrails. Once you've chosen a model, see which agent gets the most out of it.

Compare AI agents →