LFM 3 1.7B
Neural Network
LFM 3 1.7B is a compact text model with 256K context, tuned for agents, function calling, and structured output
Specifications
- Input
- text, documents
- Output
- text
- Context size
- 262 144 tokens
- Max answer length
- 8 192 tokens
- Reasoning effort
- low, medium, high
- Capabilities
- reasoning, function calling
How it works LFM 3 1.7B?
LFM 3 1.7B is a compact 1.7B parameter text model with a 256K token context, fine-tuned for agent scenarios, tool calling, and structured output. It reasons before answering, maintaining logic in step-by-step tasks. Its small parameter count makes it efficient for mass and service operations. In BotHub, the model is accessible from Russia without a VPN or foreign card; payment is in rubles based on token usage, and the balance does not expire. Over 250 models are available in one window: compare answers or switch models without rewriting integrations, as BotHub provides a unified OpenAI-compatible API. For teams, we offer contracts, invoices, electronic document management, an admin panel, and limits, with traffic encrypted via AES-GCM. Typical tasks: building agents with multiple tools, generating strict JSON for service exchange, parsing long documents and logs within 256K context, data classification and extraction in pipelines, and prototyping multimodal systems.Frequently asked questions about LFM 3 1.7B
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.
This is a compact 1.7B parameter model by Liquid AI. It generates text, parses documents with a context of up to 262,144 tokens, reasons before answering, and calls functions. Suitable for lightweight and mass tasks: rewriting fragments, extracting data from files, chatting, or processing requests within a service where speed and cost are important.
The model outputs up to 8,192 tokens per response, which is roughly several pages of text. The input context is significantly larger: up to 262,144 tokens, so you can pass a large document in the request and ask for the response in parts if necessary.
Yes, the model has a reasoning mode. It explains its thought process before the final answer, making it better at tasks requiring sequential steps: analyzing conditions, calculations, and logical deductions. For simple queries, you can disable reasoning to avoid wasting tokens.
Yes, both features are confirmed: the model supports function calling and structured JSON output. This means it can be connected to external tools and databases, and you can receive answers in a readable format. This is convenient for automation and integrations via the BotHub API.
No. The model accepts text and documents as input, such as files for analysis and summarization. It does not process images, audio, or video, nor can it generate images or sound. If you need media analysis or creation, you can switch to a suitable model in BotHub.
We do not have specific data on its performance with Russian, so we cannot promise high quality. The model is compact, and for complex Russian texts, you should verify the results with a larger model. Test simple tasks like summarization or data extraction with your own examples in BotHub.