Ling 3.1 Flash
Neural Network
Ling 3.1 Flash is a hybrid reasoning model by inclusionAI: MoE architecture, 25B active parameters out of 560B, access via BotHub API
Specifications
- Input
- text, documents
- Output
- text
- Context size
- 1 000 000 tokens
- Max answer length
- 262 144 tokens
- Reasoning effort
- low, medium, high
- Capabilities
- reasoning, function calling
Prompt cost
(per 1M tokens)
Answer cost
(per 1M 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: 'ling-3.1-flash',
});
}
// Async - Text generation
async function main() {
const stream = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'ling-3.1-flash',
stream: true
});
for await (const chunk of stream) {
const part: string | null = chunk.choices[0].delta?.content ?? null;
}
}
main();
How it works Ling 3.1 Flash?
Ling 3.1 Flash is a language model by inclusionAI built on a mixture-of-experts scheme: out of 560 billion parameters, about 25 billion are activated per request, making the heavy architecture efficient. The hybrid reasoning mode allows the model to either answer immediately or expand a chain of thought when a task requires sequential analysis. In BotHub, the model is accessible without a VPN or foreign card: you pay with a Russian card in rubles and only for tokens actually consumed, which do not expire. Nearby in the same window are over 250 other neural networks to switch between and compare answers, and a unified OpenAI-compatible API eliminates the need to rewrite integrations when changing models. Traffic is encrypted via AES-GCM, chats are not saved, and for companies, there is a contract, invoice, EDI, and admin panel with limits. Use Ling 3.1 Flash for solving logical and mathematical problems, working with and debugging code, analyzing work documents, preparing analytical summaries, and as a backend for a multimodal service.Frequently asked questions about Ling 3.1 Flash
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.
ling-3.1-flash is a text model by inclusionai: it writes and edits texts, analyzes code, helps with refactoring and debugging, and solves logic and STEM problems. It accepts documents as input, making it suitable for analyzing long reports, specifications, and correspondence. Context is 262,144 tokens.
In a single response, the model outputs up to 32,768 tokens. This is enough for a voluminous article, a detailed document analysis, or a large code module in its entirety, without splitting it into parts. If you need more, just ask it to continue — the 262,144-token context will retain the entire previous conversation.
Yes, the model has a reasoning mode: before the final answer, it breaks down the task step-by-step. This significantly helps in mathematics, logic, code analysis, and document parsing, where it is important not to guess the answer but to derive it sequentially.
Yes, ling-3.1-flash supports function calling and structured JSON output — the model selects the tool, inserts arguments, and returns the answer in the specified schema. Through the unified OpenAI-compatible BotHub API, it is easy to integrate into agents and pipelines, and if necessary, switch to another model without rewriting the integration.
Besides text, the model accepts documents — you can upload a file and ask it to summarize it, compare sections, or extract data into a table. The reception of images, audio, and video is not noted in our data, so for such tasks, it is more convenient to choose a multimodal model nearby in the same window in BotHub.
We do not have data on the quality of performance with the Russian language, so we will not make any promises. The easiest way to check is to do it yourself: give the model your real task in BotHub, compare the answer with other models in the same window, and pay only for the tokens consumed — no subscription is needed.