Gemini 2.5 Flash Lite Preview is an exact model identifier used for Google’s Gemini 2.5 Flash-Lite preview release, with the public preview model code shown as gemini-2.5-flash-lite-preview-09-2025. Google describes this preview as available in Google AI Studio and Vertex AI.
Google’s documentation for the preview lists supported input types as text, image, video, audio, and PDF, with text output.
| Developer | |
|---|---|
| Model ID | google/gemini-2.5-flash-lite-preview |
| Release Date | September 8, 2025 |
| Context Window | 1,000,000 tokens (~1M) |
| Max Output Tokens | 1,048,576 tokens |
| Modalities | Image, Text → Text |
| Capabilities | parallel tool calls, reasoning, streaming, structured (JSON schema) output, tool/function calling, vision |
Model code
The preview model code is gemini-2.5-flash-lite-preview-09-2025.
Supported data types
Inputs include text, image, video, audio, and PDF; output is text.
Supported capabilities
Google lists caching, code execution, file search, function calling, search grounding, structured outputs, thinking, and URL context as supported.
Consumption options
Google lists Batch API as supported, while Flex inference and Priority inference are not supported for the preview.
Preview positioning
Google introduced the preview as part of the Gemini 2.5 family and described it as focused on efficiency and performance in high-volume workloads.
Multimodal input workflows
The model accepts text, image, video, audio, and PDF inputs, making it suitable for mixed-format processing tasks.
Tool-augmented applications
Google lists function calling, code execution, file search, search grounding, and structured outputs as supported, which fits agentic and workflow-driven applications.
Low-latency or high-volume usage
Google positions Flash-Lite as optimized for performance and efficiency in high-volume workloads.
Gemini 2.5 Flash-Lite
Google’s documentation identifies the stable Gemini 2.5 Flash-Lite model as separate from the preview version.
Gemini 2.5 Flash
Google’s announcement groups the Flash-Lite preview alongside Gemini 2.5 Flash as part of the updated Gemini 2.5 release family.
Sources: https://ai.google.dev/gemini-api/docs/models/gemini-2.5-flash-lite · https://developers.googleblog.com/en/continuing-to-bring-you-our-latest-models-with-an-improved-gemini-2-5-flash-and-flash-lite-release/ · https://ai.google.dev/gemini-api/docs/models/gemini-2.5-flash-lite-preview-09-2025
Measured on the AI/ML API gateway across live traffic. Figures are medians unless stated. Last updated 2026-10-12.
High-volume classification is a documented fit for Gemini 2.5 Flash-Lite Preview, including classification tasks at scale and other high-throughput workloads.
Simple data extraction is another stated use case, especially when the task is lightweight, structured, and needs low latency.
Translation is explicitly called out as a good match, with the model described as well suited to translation-heavy workflows and improved translation quality in the preview version.
Extremely low-latency applications are also a stated fit, particularly where budget and speed are the main constraints.
import requests r = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={"Authorization": "Bearer " + AIMLAPI_KEY}, json={ "model": "google/gemini-2.5-flash-lite-preview", "messages": [ { "role": "user", "content": "Hello!" } ] }, ) print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/chat/completions", { method: "POST", headers: { Authorization: `Bearer ${process.env.AIMLAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ "model": "google/gemini-2.5-flash-lite-preview", "messages": [ { "role": "user", "content": "Hello!" } ] }), }); console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/chat/completions \ -H "Authorization: Bearer $AIMLAPI_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"google/gemini-2.5-flash-lite-preview","messages":[{"role":"user","content":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
Preview models will typically have billing enabled and may come with more restrictive rate limits, and this exact model is documented with a 500,000,000 token rate limit.
| Benchmark | Score | What it measures | Source | Retrieved |
|---|---|---|---|---|
| Intelligence | 10.4 | Composite score across standardised reasoning, knowledge and problem-solving evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Math | 68.7 | Composite score across standardised mathematics evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Humanity's Last Exam | 6.4% | Expert-level questions across many domains | Source | September 21, 2026 |
| GPQA Diamond | 70.2% | Google-proof graduate science questions (hardest subset) | Source | September 21, 2026 |
| AIME 2025 | 50.1% | Competition mathematics (AIME), 2025 | Source | September 21, 2026 |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
Gemini 2.5 Flash Lite Preview This page | Agentic workflows and structured output | |||
| Reasoning + agents | ||||
| Reasoning + agents | ||||
| Reasoning + agents |
Gemini 2.5 Flash Lite Preview has a 1,000,000 tokens context window and can return up to 1,048,576 tokens.
Gemini 2.5 Flash Lite Preview takes image, text as input and returns text.
Use google/gemini-2.5-flash-lite-preview as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.
Gemini 2.5 Flash Lite Preview became available on September 9, 2025.
Gemini 2.5 Flash Lite Preview is priced at input $0.0975 / 1M tokens, output $0.39 / 1M tokens, cached input $0.0975 / 1M tokens.
Yes, Gemini 2.5 Flash Lite Preview can stream responses as they are generated.
Yes, Gemini 2.5 Flash Lite Preview accepts image input alongside text.
Gemini 2.5 Flash Lite Preview was built by Google.
Yes, it supports tools including parallel tool calls and structured output.
