Gemini 2.5 Flash-Lite
Google · Released 22 Jul 2025 · gemini-2.5-flash-lite
$0.10
Input price / 1M tokens
$0.40
Output price / 1M tokens
1.05M
Context window
#11 of 71
297
Output tokens / second
#3 of 61
6.9%
Humanity’s Last Exam (no tools)
#23 of 23
66.7%
GPQA Diamond
#26 of 27
51%
MMMU-Pro
#12 of 12
34.3%
LiveCodeBench
#6 of 7
Summary
Gemini 2.5 Flash-Lite is a proprietary model from Google, released on 22 Jul 2025.
Its 1.05M-token context window ranks #11 of 71.
Measured output speed is 297 tokens per second, #3 of 61.
Smallest, most cost-efficient Gemini 2.5 model for high-volume classification, extraction and low-latency work. Closed to new users.
Benchmarks · 6 reported
| Benchmark | Score | Bar | Rank | Source |
|---|---|---|---|---|
| Humanity’s Last Exam (no tools)no tools, dynamic thinking | 6.9% | #23 of 23 | Google Gemini 3.1 Flash-Lite evaluation report (Gemini 2.5 Flash-Lite column, Mar 2026)Self-reported | |
| GPQA Diamondno tools, dynamic thinking | 66.7% | #26 of 27 | Google Gemini 3.1 Flash-Lite evaluation report (Gemini 2.5 Flash-Lite column, Mar 2026)Self-reported | |
| MMMU-Prono tools | 51% | #12 of 12 | Google Gemini 3.1 Flash-Lite evaluation report (Gemini 2.5 Flash-Lite column, Mar 2026)Self-reported | |
| MMMLU | 84.5% | – | Google Gemini 3.1 Flash-Lite evaluation report (Gemini 2.5 Flash-Lite column, Mar 2026)Self-reported | |
| LiveCodeBenchcode generation, 175 UI problems dated 1 Jan 2025 to 1 May 2025 | 34.3% | #6 of 7 | Google Gemini 3.1 Flash-Lite evaluation report (Gemini 2.5 Flash-Lite column, Mar 2026)Self-reported | |
| MRCR v2 (8-needle)128k average (5.4% at 1M pointwise) | 30.6% | #13 of 13 | Google Gemini 3.1 Flash-Lite evaluation report (Gemini 2.5 Flash-Lite column, Mar 2026)Self-reported |
Self-reported means the lab ran the test itself. A rank appears only where several labs report the same version of a benchmark.
Similar models
Models with the most benchmarks in common with Gemini 2.5 Flash-Lite, and the closest scores.
Humanity’s Last Exam (no tools)
#23 of 23Expert questions, no search or code
GPQA Diamond
#26 of 27Graduate-level science questions
MMMU-Pro
#12 of 12Hard questions that need an image
LiveCodeBench
#6 of 7Fresh competitive programming problems
Compare Gemini 2.5 Flash-Lite side by side
Pre-filled with the two closest models. Swap any of them.
Sources
- price, cache, batch, free tier: ai.google.dev/gemini-api/docs/pricing, checked 2 Oct 2026
- context, max output, knowledge cutoff, modalities, access limit: ai.google.dev/gemini-api/docs/models/gemini-2.5-flash-lite, checked 2 Oct 2026
- release date, no shutdown date: ai.google.dev/gemini-api/docs/deprecations, checked 2 Oct 2026
- benchmarks: storage.googleapis.com/deepmind-media/gemini/gemini_3-1_flash-lite_model_evaluat, checked 2 Oct 2026
- output speed: artificialanalysis.ai/models/gemini-2-5-flash-lite, checked 2 Oct 2026
Building on Gemini 2.5 Flash-Lite?
Get the architecture right before the bill arrives.
We size caching, routing and fallbacks for your workload on a free call, and tell you where a cheaper model is good enough.