How it works Inkling?
Inkling is an open-weights multimodal model by Thinking Machines Lab based on a mixture-of-experts architecture: out of 975 billion parameters, about 41 billion are activated per request, combining the knowledge of a large network with efficient computation. It was designed for universal reasoning, code generation and analysis, and agentic systems with tool calling, making Inkling suitable for tasks requiring step-by-step reasoning. In BotHub, the model is accessible without a VPN or foreign card: pay with a Russian card in rubles only for consumed tokens that do not expire. Compare answers from over 250 models in one window without rewriting integrations: BotHub offers a unified API for all models. Chats are encrypted, and companies have access to contracts, invoices, electronic document management, and an admin panel with limits. Useful for refactoring code, debugging trace errors, building agents with function calling, drafting technical documentation, or analyzing images alongside text prompts.Frequently asked questions about Inkling
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.
Inkling:free by thinkingmachines is a text model with pre-response reasoning. It analyzes documents and images, writes and edits code, and assists with analytics and STEM tasks. The 262,144-token context window allows handling large files and long conversations in their entirety without splitting them into fragments.
The output limit, according to our data, is 262,144 tokens, the same as the context. This means you can get a long document, a large code module, or a detailed analysis in a single pass without manually stitching results together.
Yes, the reasoning mode is confirmed. The model processes the task step-by-step before formulating an answer, which significantly helps with math, logic, code debugging, and multi-step scenarios. You pay only for tokens used; no subscription is required.
Yes, the model supports function calling and structured JSON output, allowing it to be used in agentic scenarios and connected to external tools. BotHub uses a unified OpenAI-compatible API, so you can switch models without rewriting integrations.
The model accepts images and documents as input: you can send screenshots, diagrams, PDFs, or tables and ask it to analyze the content. We have no data on audio or video support; for such tasks, BotHub offers specialized models.
We have no data on language capabilities, so we cannot make promises: test the model with your own real-world queries. It's fast—access from Russia works without a VPN or foreign card, payment is in rubles, and you can compare answers from other models in the same window.