Gemini 2.5 Flash Lite
Neural Network
Gemini 2.5 Flash Lite is a lightweight Google reasoning model designed for minimal latency and high generation speed.
Specifications
- Input
- text, images, documents
- Output
- text
- Context size
- 1 048 576 tokens
- Max answer length
- 65 535 tokens
- Reasoning effort
- low, medium, high
- Capabilities
- reasoning, function calling, Batch API
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
Image prompt
(per 1K tokens)
import OpenAI from 'openai';
const openai = new OpenAI({
apiKey: '<your bothub access token>',
baseURL: 'https://openai.bothub.chat/v1'
});
// Sync - Text generation
async function main() {
const chatCompletion = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'gemini-2.5-flash-lite',
});
}
// Async - Text generation
async function main() {
const stream = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'gemini-2.5-flash-lite',
stream: true
});
for await (const chunk of stream) {
const part: string | null = chunk.choices[0].delta?.content ?? null;
}
}
main();
How it works Gemini 2.5 Flash Lite?
Gemini 2.5 Flash Lite is a lightweight reasoning model from Google's Gemini 2.5 family, built for speed: ultra-low latency, increased throughput, and cost-effective operation for high-volume requests. Compared to the previous version, tokens are generated faster, and overall performance is higher. In BotHub, it is accessible from Russia without a VPN or foreign card, and you pay in rubles with a Russian card based on actual usage—for consumed tokens, without a mandatory subscription, and tokens do not expire. Over 250 neural networks are available in the same window: it's easy to compare answers and switch models without changing tools. The unified BotHub API connects Gemini 2.5 Flash Lite to your service, and switching models won't require rewriting the integration; chats are encrypted via AES-GCM and not saved. Companies have access to contracts, invoices, EDI, an admin panel, and employee limits. Simple scenarios: support chatbots with instant responses, streaming processing and classification of large text arrays, summaries and drafts, in-product interface prompts, and fast iterations for prompt debugging.Frequently asked questions about Gemini 2.5 Flash Lite
You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.
The model works with text, documents, and images: it parses long files and answers questions about them, writes and edits code, and helps with reasoning and STEM tasks. The context is 1,048,576 tokens, so large documentation or an entire chat history will fit into a request.
A single response yields up to 65,535 tokens — that's dozens of pages. It's enough for a detailed document analysis, a voluminous translation, or an entire code module. If the material doesn't fit, just ask it to continue: the 1,048,576 token context is enough to hold the entire dialogue.
Yes, reasoning before answering is supported: before providing a result, the model runs a chain of thought. This is noticeable in math, logic, data analysis, and code debugging, where a task needs to be broken down into steps. The more complex the request, the more noticeable the effect.
Yes, both function calling and structured JSON output are available. You describe the tools, the model decides when to apply them, and returns the answer in the specified schema — this is convenient for agents, bots, and integrations. Connection is via the unified OpenAI-compatible BotHub API.
According to our data, text, images, and documents are accepted as input: the model reads screenshots, diagrams, tables, scans, and PDFs natively, without prior conversion. We have no notes regarding audio and video, so you should test this scenario with your own file directly in the chat.
There is no specific note about languages in our data, so we won't make any promises — it's more reliable to test it with your own text. It's easy to do: in BotHub, the model is available without a VPN or foreign card, payment is in rubles based on actual usage, and other models are available in the same window for comparison.