SOTA LLM Pricing Comparison

API KEY is needed to chat PDF in PapersGPT for online LLMs

Almost all the mainstream LLMs(large language models) are supported in PapersGPT, and LLM API KEY should be provided or bought by the users themselves. While which LLM is the most suitable for you, which is more cost-effective, the following gives you some reference information.

What's the token?

Before making a price comparison, we first need to figure out what a token is. In the field of Artificial Intelligence and Natural Language Processing, Token is the basic unit of text after segmentation. The number of English words contained in a Token is not fixed. In English, common short words such as "the" "and" are a Token, while longer words such as "hesitation" are also a Token. As a rough estimate, on average, an English Token may correspond to 3–5 letters.

What's the SOTA(State Of The Art) Model?

The "SOTA LLM model" is a constantly moving target. And there isn't one single, universally declared "SOTA" (State-of-the-Art) LLM that definitively outclasses all others across every single metric. Instead, what's considered SOTA depends heavily on specific tasks, benchmarks, and evaluation criteria. Different models excel in different areas. However, Until August 2025, the most widely used and recognized top-tier smart models include:

* GPT 5，o1/2/3/4, GPT 4o(OpenAI): This is arguably the most widely recognized and influential SOTA model. It excels in a broad range of tasks, including reasoning, coding, creative writing, complex instruction following, and general knowledge. Its capabilities are vast.

* Gemini(Google): Leveraging Google's powerful technology product development system, massive computing chips, and vast amounts of high-quality data, Gemini Pro 2.5 quickly swept the top of major model evaluation rankings. It developed very rapidly, was very affordable, and had a series of free trial plans, grabbing a large market share.

* Claude(Anthropic)：A very professional model that emphasizes practical usability. In some complex reasoning, especially in the field of vibe coding, it is the best and its effect far exceeds other models on the market.

* Grok 4(xAI)：Thanks to the financial resources, strong appeal and execution of the world's richest man, Elon Mask, the Grok 4 series of models has been a huge success, and is basically at the top of the rankings of major model reviews.

The smartest second-tier LLMs, approaching the first tier and developing rapidly, include:

* DeepSeek: In early 2025, thanks to the successful launch of DeepSeek V3/R1, DeepSeek catapulted from obscurity to become a global leader in large-scale model development. Its primary selling point is its cost-effectiveness, significantly lower than competing products for comparable performance. Its models are also highly intelligent, approaching state-of-the-art performance. Furthermore, their fully open-source model allows for easy customization and deployment, significantly lowering the barrier to entry for large-scale model adoption.

* Mistral: Come from Europe, known for its multilingualism and open source nature. Its latest Medium series has achieved very good results in some reviews, and it has unique support for minority languages in European countries, such as French, German, and Italian.

* Kimi: kimi-k2 is a new open source model that emerged in July 2025. It has achieved very good results in a series of evaluations, especially in the use scenarios of coding and agent, where its advantages are more obvious.

There are many evaluation rankings for LLMs. Here recommending lmarena.ai. Its main feature is to use actual and relatively subjective manual evaluation as the standard. In actual conversations and usage scenarios, humans manually score the participating black-box LLMs.

Which is the most cost-effective Model?

Currently, in the paper reading scenario, most SOTA or near-SOTA models perform very well, unless the paper is very long or contains a large number of charts, etc. Therefore, it is very necessary to choose a LLM with good cost-effectiveness. Below is the latest SOTA LLM API pricing I compiled until August 2025.

The final decision is yours

Although there are various LLMs on the market, you don't need to worry about choosing. The best and most affordable LLM is determined by its practical use. You only need to choose one in PapersGPT and use it. If you think the effect is not good or the price is too high, you can switch to another model with one click. It is very convenient.