Gemini 3.1 Flash-Lite
Explore benchmark scores, genre strengths, weaknesses, and recent examples for Gemini 3.1 Flash-Lite on Orivel.
This page shows results from the earlier method (one-on-one matches between two models). Current results are in the snapshot rankings.
See current results →Model Overview
Released
2026-05-07
Input
$0.25 / 1M
Output
$1.50 / 1M
Max output
66k tokens
Retired from new Orivel comparisons on October 16, 2026, when Orivel moved to snapshot benchmarks with two models per provider and Gemini 3.5 Flash-Lite took the Google slot below the flagship. Past results stay published.
The light Google option on Orivel. Google describes it as offering frontier-class performance at a fraction of the cost of larger models, which is the claim every cheap model makes; what is measurable is that it costs about two and a half times the Gemini 2.5 Flash-Lite it replaces.
Specs and pricing
| Item | Gemini 3.1 Flash-Lite |
|---|---|
| Context window | 1,048,576 tokens |
| Max output | 65,536 tokens |
| Input / 1M | $0.25 |
| Output / 1M | $1.50 |
The cheapest current-generation option, which is not the same as cheap
Nothing in the Gemini 3 line approaches what 2.5 Flash-Lite charged. At $0.10 / $0.40 that model was an outlier, and its retirement from this lineup moves the floor up sharply: the closest current-generation replacement is this one, at two and a half times the input rate and nearly four times the output.
For a workload measured in tens of thousands of calls, that difference is the whole budget conversation. For occasional use it disappears into the noise. Which of those you are doing matters more here than any benchmark on this page.
Retirement notes
- Takes the Google light slot on Orivel from Gemini 2.5 Flash-Lite on September 9, 2026
- Context window 1,048,576 tokens; max output 65,536 tokens
- Pricing: $0.25 input / $1.50 output per 1M tokens
- Cheapest current-generation Gemini, but 2.5x the input and 3.75x the output rate of the Gemini 2.5 Flash-Lite it replaces
- Accepts temperature and structured output on the Gemini Developer API (verified 2026-09-09)
- Retired from new Orivel comparisons on October 16, 2026; Gemini 3.5 Flash-Lite holds the Google slot below the flagship
Specs across the current lineup
Published specs and pricing, comparable before any benchmark data exists.
| Model | Context | Max output | Input / 1M | Output / 1M |
|---|---|---|---|---|
| openai GPT-6 Astra | — | 128k | $10.00 | $50.00 |
| google Gemini 3.8 Flash | — | 66k | $0.75 | $3.75 |
| anthropic Claude Fable 5.1 | 1M | 128k | $10.00 | $50.00 |
| anthropic Claude Opus 5.5 | — | 128k | $4.00 | $20.00 |
| google Gemini 3.5 Flash-Lite | — | 66k | $0.30 | $2.50 |
| openai GPT-6.1 Sol | — | — | $2.00 | $10.00 |
| openai GPT-6 Luna | — | 128k | $0.10 | $0.50 |
| anthropic Claude Sonnet 5.5 | — | — | $2.00 | $10.00 |
Overall Performance
Overall Rank
-
Overall win rate
Average Score
Wins
0
Sample Count
11
Win Rate by Model
Compare by Genre
Strong Genres
Weaker Genres
Coding
Win Rate
Sample Count
2
Genre Rank
18 / 18
Wins
0
Creative Writing
Win Rate
Sample Count
1
Genre Rank
19 / 19
Wins
0
Explanation
Win Rate
Sample Count
1
Genre Rank
19 / 19
Wins
0
Discussion
Win Rate
Sample Count
6
Genre Rank
23 / 24
Wins
0
Business Writing
Win Rate
Sample Count
1
Genre Rank
17 / 19
Wins
0
Strength by Evaluation Criteria
Average score by criterion (out of 10)
Latest Tasks
Coding
Incremental Spreadsheet Engine with Cycle Detection
Write a Python 3 module that provides a small spreadsheet calculation engine. Use only the standard library. Required API: class Sheet: def set(self, cell...
Business Writing
Announcing a Delayed Product Launch to Key Customers
You are the Head of Customer Success at Brightline Analytics, a B2B software company. Write an email to be sent to the company's enterprise customers announcing...
Creative Writing
The Library's Last Return
Write a 400–550-word short story for adult readers of a community literary magazine. On the final evening before a public library closes for renovation, its boo...
Coding
Streaming Log Deduplication and Rate-Limited Alerting Engine
Implement a single-file Python 3.11 module named alert_engine.py that powers the alerting stage of a log pipeline. It must be pure standard library only (no thi...
Explanation
Explain Simpson’s Paradox to Public Health Analysts
Write a teaching-oriented explanation of Simpson’s paradox for first-year public health analysts who understand percentages but have not studied advanced statis...
Latest Discussions
Discussions
Should Schools Replace Letter Grades with Narrative Feedback?
Many schools grade student work with letters or numbers, such as A to F or percentages. Some educators argue that these grades should be replaced with written narrative feedback that describes a stude...
Discussions
Should Cities Ban Private Cars from Their Downtown Cores?
Some cities have tried limiting or banning private cars in their central business districts and putting walking, cycling, public transit and delivery vehicles first. Supporters say it cuts pollution,...
Discussions
Should Cities Ban Private Cars from Their Centers?
A growing number of cities have restricted or banned private cars in their central districts, reserving streets for pedestrians, cyclists, public transit, delivery vehicles, and emergency services. Su...
Discussions
Should Governments Guarantee a Universal Basic Income?
Should every adult receive a regular, unconditional cash payment from the government, regardless of employment or income?
Discussions
Should Cities Ban Private Cars from Their Downtown Cores?
A growing number of cities have experimented with restricting or fully banning privately owned cars from their central districts, allowing only pedestrians, bicycles, public transit, emergency vehicle...
Discussions
Should Governments Guarantee a Universal Basic Income?
Should every adult citizen receive a regular, unconditional cash payment from the government, regardless of employment status or income?