We analyzed 15+ LLMs and their pricing and performance. LLM API pricing can be complex and depends on your preferred usage. If you plan to use:
- The chat user interface, see all major LLM subscription plans
- APIs, see LLMs ranked by their performance and type your volume needs in terms of tokens to see the exact pricing:
| Model | Input Price |
Output Price |
Context Length |
Max Output Tokens |
Arena Score |
|---|---|---|---|---|---|
| Google Gemini-2.5-Pro | $2.50 | $15.00 | 1,000k | 65k | 1,446 |
| OpenAI o3 | $10.00 | $40.00 | 200k | 100k | 1,413 |
| OpenAI gpt-4o | $2.50 | $10.00 | 128k | 16k | 1,408 |
| xAI Grok-3-Preview | $3.00 | $15.00 | 131k | n/a | 1,403 |
| OpenAI gpt-4.5 | $75.00 | $150.00 | 128k | 16k | 1,398 |
| DeepSeek DeepSeek-R1 | $0.55 | $2.19 | 64k | 8k | 1,358 |
| Google Gemini-2.0-Flash-001 | $0.10 | $0.40 | 1,000k | 8k | 1,355 |
| OpenAI o1-2024-12-17 | $15.00 | $60.00 | 200k | 100k | 1,350 |
| Google Gemma-3-27B-it | n/a | n/a | 128k | 8k | 1,341 |
| Alibaba Qwen2.5-Max | $1.60 | $6.40 | 32k | 8k | 1,341 |
| OpenAI o1-preview | $15.00 | $60.00 | 128k | 32k | 1,335 |
| OpenAI o3-mini-high | $1.10 | $4.40 | 200k | 100k | 1,325 |
| DeepSeek DeepSeek-V3 | $0.27 | $1.10 | 64k | 8k | 1,318 |
| Alibaba Qwen-Plus-0125 | $0.40 | $1.20 | 131k | 8k | 1,310 |
| Cohere Command A | $2.50 | $10.00 | 256k | 8k | 1,306 |
| OpenAI o3-mini | $1.10 | $4.40 | 200k | 100k | 1,305 |
| OpenAI o1-mini | $1.10 | $4.40 | 128k | 65k | 1,304 |
| Anthropic Claude 3.7 Sonnet | $3.00 | $15.00 | 200k | 128k | 1,290 |
Hover over model names to see their full names and over headers to see explanations about the columns.
Ranking: The ranking of models is based on the Overall Arena Score in the Chatbot Arena LLM leaderboard.1
When the official Grok 3 API is released and Gemini’s experimental models exit their experimental phase, the missing values will be populated with the official values.
Rate limits control Google API request frequency for free tiers.2
You can check the hallucination rates and reasoning performance of top LLMs in our benchmarks.
Understanding LLM Pricing
Tokens: The Fundamental Unit of Pricing
While providers offer a variety of pricing structures, per-token pricing is the most common. Tokenization methods differ across models; examples include:
- Byte-Pair Encoding (BPE): Splits words into frequent subword units, balancing vocabulary size and efficiency.4
- Example: “unbelievable” → [“un”, “believ”, “able”]
- WordPiece: Similar to BPE but optimizes for language model likelihood, used in BERT.5
- Example: “tokenization” → [“token”, “##ization”]. “token” is a standalone word; “##ization” is a suffix.
- SentencePiece: Tokenizes text without relying on spaces, effective for multilingual models like T5.6
- Example: “natural language” → [” natural”, ” lan”, “guage”] or [” natu”, “ral”, ” language”].
Please note that the exact subwords depend on the training data and BPE/WordPiece process. To better understand these tokenization methods, watch the video below:
After grasping tokenization, an average price can be estimated based on project token length. Table 2 outlines token ranges by content type—such as UI prompts, email snippets, marketing blogs, detailed reports, and research papers—while noting that token counts vary across models. Once a model is chosen, its tokenizer can be used to have an idea of the average token count for the content.
| Content Type | Word Count Range (words) | Token Count Range (tokens) | Typical Enterprise Use Cases |
|---|---|---|---|
| Sentence | 10–20 | 15–35 | UI prompts, notifications, chatbot responses |
| Paragraph | 75–150 | 100–225 | Email snippets, product descriptions, help texts |
| Short Article | 400–600 | 520–900 | Marketing blogs, press releases, case studies |
| Long Article | 900–1,100 | 1,200–1,650 | Detailed reports, whitepapers, internal knowledge bases |
| Research Paper | 4,500–5,500 | 5,850–8,250 | Academic publications, R&D documents, technical whitepapers |
Table 2. Typical content types, their size ranges, and enterprise considerations (ranges are estimates and may vary).
Context Window Implications
Awareness of the context window concept is another crucial factor to consider regarding pricing. Here, it is essential to ensure that the total number of tokens from both the input and output does not exceed the context window/length.
If the total exceeds the context window, it may lead to the truncation of the excess output, as shown in Figure 2. Therefore, the output may not be as expected. It is important to note that tokens generated during the reasoning process are also counted within this limitation.
Max Output Tokens
This is an important parameter in Large Language Models (LLMs) to achieve the desired output and manage costs effectively. While many documentations mention that it can be adjusted using the max_tokens parameter, it is crucial to review the documentation of the specific API being used to identify the correct parameter. It should be adjusted according to the specific needs:
If set too low: It may result in incomplete outputs, causing the model to cut off responses before delivering the full answer.
If set too high: Depending on the temperature (a parameter that controls response creativity) setting, it can lead to unnecessarily verbose outputs, longer response times, and increased cost.
Therefore, it is a parameter that requires careful consideration to optimize resource usage while balancing output quality, cost, and performance.
| Content Type | Input Prompt Example | Input Token Count* | Assumed Output Token Count* |
|---|---|---|---|
| Sentence | “Generate a friendly notification message reminding users to complete their profile within the app.” | 15 | 25 |
| Paragraph | “Write a concise email snippet announcing the launch of our new product feature, highlighting its key benefit.” | 19 | 162 |
| Short Article | “Create a short blog post explaining how our new software solution improves remote team productivity.” | 18 | 710 |
| Long Article | “Draft a comprehensive whitepaper outlining the impact of AI on the future of supply chain management, including real-world case studies.” | 24 | 1,425 |
| Research Paper | “Write a comprehensive full-length research paper on the application of machine learning algorithms in geological data analysis, covering background, literature review, theoretical framework, methodology, results, discussion, and referencing recent studies.” | 26 | 7,050 |
Table 3. Example input prompts and estimated token counts per content type.
*This assumes that each model produces responses with an equal number of output tokens—although the token count for both input and output may vary depending on each model’s tokenization, the number has been kept constant here for each model.
The LLM API price calculator can be used to determine the total cost per model when generating the content types from Table 2 via the API using the sample prompts provided in Table 3, as well as for calculating costs for custom cases beyond the suggested content types.
LLM API Price Calculator
You can calculate your total cost by filling out these 3 values below and sort the results by input cost, output cost, total cost, or alphabetically in increasing or decreasing order:
The default ranking is based on the total cost.
Comparing LLM Subscription Plans
Non-technical users may prefer to use the UI rather than the API, here is the UI pricing:
| Plan | Price | Key Features | Limitations |
|---|---|---|---|
| Microsoft Copilot/Free | $0 | Basic Microsoft app integration Multiple devices and platforms 15 boosts per day | Limited credits for AI usage (includes designer only) Preferred model access only during non-peak times Limited Copilot Voice usage |
| Microsoft Copilot/Pro | $20/month | Preferred model access 100 boosts per day Full Microsoft 365 integration Early feature access All Microsoft app support | Individual use only |
| Google Gemini/Basic | $0 | 2.0 Flash model access Basic writing & images Google apps integration Voice conversations | Basic model only, Limited features |
| Google Gemini/Advanced | $19.99/month | 2.0 Pro experimental model Deep research capability 1500-page document analysis 2TB storage Custom AI experts (Gems) Code smarter | |
| Mistral AI/Free | $0 | Web browsing & news Basic file analysis Image generation Flash answers | Training data usage, Limited capabilities |
| Mistral AI/Pro | $14.99/month | Unlimited web browsing Extended analysis Opt-out data sharing Dedicated support | Individual use only |
| Mistral AI/Team | $19.99 per user / month billed annualy $24.99 per user / month billed monthly | Central billing API credits Data excluded from training Advanced features | Minimum 2 members |
| Mistral AI/Enterprise | Custom | Secure deployment in your environment Enhanced support Granular admin controls Detailed analytics | |
| OpenAI/Free | $0 | GPT-4o mini access Standard voice Limited uploads Basic images | Usage caps, Basic models only |
| OpenAI/Plus | $20/month | Extended limits Advanced voice Beta features Limited GPT-4 | Individual use, Usage must follow policies |
| OpenAI/Pro | $200/month | Unlimited access to o1, o1-mini, GPT-4o, and voice (audio only) Higher limits for video and screensharing in voice Access to o1 Pro mode Extended access to Sora Access to Operator research preview (U.S. Only) | Usage must be reasonable and comply with their policies |
| OpenAI/Team | $25 per user / month billed annualy $30 per user / month billed monthly | Higher message limits than Plus on GPT-4, GPT-4o, and tools like DALL·E Standard and advanced voice mode Admin console for workspace management Team data excluded from training by default | Minimum 2 members |
| OpenAI/Enterprise | Custom | High speed access to GPT-4, GPT-4o, GPT-4o mini, and tools like DALL E Expanded context window Enterprise data excluded from training by default & custom data retention windows Admin controls, domain verification, and analytics, Enhanced support | |
| Claude.ai/Free | $0 | Web/mobile access Basic analysis One latest model Document upload | Limited usage, Basic features only |
| Claude.ai/Pro | $18/month billed annualy $20/month billed monthly | Claude 3.5 Sonnet & Opus access Projects organization Increased usage Early features | Individual use only |
| Claude.ai/Team | $25 per user / month billed annualy $30 per user / month billed monthly | Central billing Collaboration tools Expanded usage Admin features | 5 member minimum |
| Claude.ai/Enterprise | Custom | Expanded context window Single sign-on (SSO) and domain capture Role-based access System for Cross-domain Identity Management (SCIM) Audit logs Data source integrations |
Using Multiple Language Models
A tool like OpenRouter allows the same prompt to be sent simultaneously to multiple models. The responses, token consumption, response time, and pricing can then be compared to determine which model is most suitable for the task.
Benefits and Challenges
- Increased Adaptability and Efficiency: Orchestration enhances responsiveness, allowing for real-time assessment of model efficiency, leading to the identification of a cost-effective model and potential savings.
- Prompt Sensitivity and Optimization: Identical prompts can elicit vastly different outputs across models, necessitating prompt engineering tailored to each model to achieve desired results, adding to development and maintenance complexity.
FAQ
What is LLM API Pricing?
Accessing Large Language Models (LLMs) via an Application Programming Interface (API) grants you remote access to AI models. This access is subject to a fee, often called an “API fee,” charged by the service provider. This fee is a critical consideration when integrating LLMs into your applications. It essentially represents the cost associated with each query, request, or task performed through the provider’s API. Because pricing structures can vary widely – based on factors like token usage, API call volume, feature utilization, or subscription models – understanding how providers calculate these costs is essential. With this knowledge, you can make well-informed decisions, selecting the LLM model and provider that best balance your performance needs, desired functionality, and budgetary limitations.
Why is LLM API pricing complex?
LLM API pricing can be complex due to factors like token consumption, context length, and model choice. Tokenization procedures vary across models, with some using Byte-Pair Encoding (BPE), WordPiece, or SentencePiece—each influencing how text is split into tokens and impacting cost efficiency. Understanding these differences helps optimize API usage and pricing.
What factors determine the cost of using a large language model (LLM)?
LLM costs are primarily determined by token usage (both input and output), API call volume, and the specific pricing model (e.g., per-token, subscription).
How can I compare pricing across different LLM models?
Compare input and output token prices, context window limits, and any additional fees. Tools like OpenRouter allow you to send the same prompt to multiple models and directly compare their results, token usage, speed, and pricing. Consider your typical content length and usage patterns to estimate overall costs.
What is the difference between input tokens and output tokens?
Input tokens are the tokens in the prompt you send to the LLM, while output tokens are the tokens in the generated response. For reasoning models, it’s important to note that tokens generated during the reasoning process itself are also counted as output tokens, impacting the final cost. Both input and output contribute to the overall cost.
How does the text volume I request affect the processing response time and overall budget when using an LLM API?
Larger text requests require more processing, increasing response time and costs. Optimize input sizes and use an LLM API pricing calculator to estimate token counts and manage your budget effectively.
What resources are available to the LLM community to support understanding and optimizing LLM pricing information?
The LLM community has developed various tools and benchmarks to help users understand and optimize LLM pricing. These resources often include calculators and comparison charts that offer insights into the power and efficiency of different models. Platforms like Hugging Face and GitHub host tools and code developed by the community to analyze model performance and costs. Many services offer community support through forums or chat features.
External Links
- 1. https://lmarena.ai/?leaderboard
- 2. Rate limits | Gemini API | Google AI for Developers.
- 3. OpenAI Platform.
- 4. [1508.07909] Neural Machine Translation of Rare Words with Subword Units.
- 5. [1810.04805] BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding.
- 6. [1808.06226] SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing.
- 7. OpenAI Platform.
- 8. OpenRouter.
Comments
Your email address will not be published. All fields are required.