Solar Pro 4
Neural Network
Solar Pro 4 is an Upstage language model with a 524K context window for agent scenarios and document-intensive tasks.
Specifications
- Input
- text, images, documents
- Output
- text
- Context size
- 524 288 tokens
- Max answer length
- 131 072 tokens
- Reasoning effort
- none, minimal, low, medium, high, extra high, max
- Capabilities
- reasoning, function calling
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
How it works Solar Pro 4?
Solar Pro 4 is an Upstage language model with a 524K context window, designed for long multi-step tasks and agent scenarios. Its strength lies in document-intensive and office processes: large materials fit into the context entirely without being split into pieces. The vendor emphasizes the model's cost-effectiveness. BotHub provides access without VPN or foreign cards; payment is made with Russian cards in rubles, based on actual usage—tokens do not expire. With over 250 models in one window, comparing responses and switching is easy, and a unified OpenAI-compatible API eliminates the need to rewrite integrations. Data is transmitted encrypted, and companies have access to contracts, invoices, electronic document management (EDO), and an admin panel with limits. Useful for analyzing multi-page contracts and reports, summarizing dozens of files, building an agent that keeps the entire task history in memory, or automating routine document work.Frequently asked questions about Solar Pro 4
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.
solar-pro4 by upstage is a reasoning text model: it reads long documents, analyzes code, helps with analytics and STEM tasks, and can call functions. It accepts not only text but also images and documents as input, making it suitable for extracting data from scans, tables, and screenshots.
The output limit is 131,072 tokens per response, and the context window holds 524,288 tokens. This is enough to get a large technical analysis, a long report, or a substantial piece of code in its entirety without splitting the task into multiple requests.
Yes, reasoning before answering is confirmed by our data. The model first builds a chain of steps and then provides the result—this significantly helps with math, code debugging, and multi-step logic. In BotHub, you pay for tokens as you use them; no subscription is required.
Yes. solar-pro4 supports function calling and structured JSON output, making it convenient for agents and pipelines. Connection is via the unified OpenAI-compatible BotHub API: you can switch to another model later without rewriting the integration.
You can upload images and documents—they are listed among the supported input formats. There is no mention of audio or video in our data, so expect it to handle text, files, and images. The model outputs text.
We do not have specific data on languages, so it is fairer to test it with your own tasks—a couple of requests will show if the model is suitable. In BotHub, it is simple: access from Russia without VPN or foreign cards, payment with a ruble card, and other models are available in the same window for comparison.