Pricing Overview
At OpenKBS, customers pay only for the resources they use, ensuring cost efficiency across a diverse range of services. This model is designed to support scalability, allowing you to adjust your usage based on current needs without facing unexpected fees or charges. For every transaction on our platform, there is a nominal fee of 1 credit, equivalent to exactly $0.00001. This micro-transaction fee is part of our commitment to maintaining a transparent and fair pricing structure, ensuring that you are billed precisely for the resources consumed.
Chat Models
AI chat models are priced based on the usage of input and output tokens, calculated per 1000 tokens. A token can be a word, part of a word, or punctuation, serving as the basic unit the model processes. Input tokens are the text prompts given to the model, while output tokens are the responses generated by the model. For streaming APIs, see the TOS for token consumption differences.
| Service Name | Vendor | Text Input | Text Output | Image Input | Context Size |
|---|---|---|---|---|---|
| GPT 4 Omni | openai | $0.00500 / 1K tokens | $0.01500 / 1K tokens | $0.00750 / image | 128K |
| GPT 4o Mini | openai | $0.00020 / 1K tokens | $0.00080 / 1K tokens | $0.00750 / image | 128K |
| GPT 4 Turbo | openai | $0.01000 / 1K tokens | $0.03000 / 1K tokens | 128K | |
| GPT 3.5 Turbo | openai | $0.00050 / 1K tokens | $0.00150 / 1K tokens | 16K | |
| Claude 3.5 Sonnet | anthropic | $0.00400 / 1K tokens | $0.02000 / 1K tokens | 200K | |
| Claude 3 Opus | anthropic | $0.01500 / 1K tokens | $0.07500 / 1K tokens | 200K | |
| Claude 3 Sonnet | anthropic | $0.00300 / 1K tokens | $0.01500 / 1K tokens | 200K | |
| Claude 3 Haiku | anthropic | $0.00025 / 1K tokens | $0.00125 / 1K tokens | 200K | |
| Llama 3.1 405B | meta | $0.00500 / 1K tokens | $0.01600 / 1K tokens | 120K | |
| Llama 3.1 70B | meta | $0.00099 / 1K tokens | $0.00099 / 1K tokens | 120K | |
| Llama 3 70B | meta | $0.00265 / 1K tokens | $0.00350 / 1K tokens | 8K | |
| Llama 3 8B | meta | $0.00040 / 1K tokens | $0.00060 / 1K tokens | 8K | |
| Mistral Large | mistral | $0.00800 / 1K tokens | $0.02400 / 1K tokens | 32K | |
| Mixtral-8x7b | mistral | $0.00045 / 1K tokens | $0.00070 / 1K tokens | 32K | |
| Mistral 7b | mistral | $0.00015 / 1K tokens | $0.00020 / 1K tokens | 32K |
Image Generation
Diffusion models, which convert text descriptions into images, are priced on a per-request basis. These models leverage advanced algorithms to interpret textual prompts and generate corresponding visual content. Pricing is standardized per request, ensuring straightforward and consistent pricing.
| Service Name | Vendor | Price | Description |
|---|---|---|---|
| Stable Diffusion 3 Medium | $0.010 / request | Image generation model by Stability AI | |
| Stable Diffusion XL 1.0 | $0.002 / request | Image generation model by Stability AI | |
| Animagine XL 3.0 | $0.004 / request | Generates high-quality anime images from textual prompts | |
| Diffusers SDXL Inpaint | $0.004 / request | A latent text-to-image diffusion model | |
| Runwayml SDXL Inpaint | $0.004 / request | A latent text-to-image diffusion model | |
| Stable Diffusion XL Inpaint | $0.004 / request | A latent text-to-image diffusion model | |
| Stable Diffusion Upscaler x4 | $0.008 / request | Upscaler model |
Embedding Models
Embeddings models are priced based on the usage of input tokens, calculated per 1000 tokens. A token can be a word, part of a word, or punctuation, serving as the basic unit the model processes. Input tokens are the text data provided to the model for generating embeddings.
| Service Name | Vendor | Text Input | Context Size |
|---|---|---|---|
| OpenAI Text Embedding v3 large | openai | $0.00013 / 1K tokens | 8K |
| OpenAI Text Embedding v3 small | openai | $0.00002 / 1K tokens | 8K |
Web Tools
Web tools encompass a range of services designed to interact with web content, including search engines, webpage readers, and more. These tools are priced based on the complexity and the computational resources required for each request.
| Service Name | Vendor | Price | Description |
|---|---|---|---|
| Google Search | $0.006 / request | Retrieves search results from Google | |
| Read Webpage | $0.0048 / minute | Extracts text from web pages | |
| GCP Text Detection | $0.002 / request | Extracts text from images | |
| Document To Text | $0.0024 / minute | Extracts text from documents | |
| Translate Text | $0.00003 / character | Provides text translation and detection |
Audio Models
Audio models offer capabilities ranging from speech-to-text conversion to generating synthetic speech. Pricing varies based on the model's complexity and the length of the audio processed.
| Service Name | Vendor | Price | Description |
|---|---|---|---|
| Whisper Large v3 | $0.0060 / minute | Extracts text from speech | |
| GCP Text To Speech | $0.00003 / character | Converts text into natural-sounding speech |
Serverless NoSQL Database
The serverless NoSQL database service is fully managed, scalable, and high-performance. Pricing is based on data storage and read/write operations. Read costs are measured in Read Units (RUs), with one RU covering items up to 4 KB. Write costs are measured in Write Units (WUs), with one WU covering items up to 1 KB.
| Service Name | Read | Write | Storage | Description |
|---|---|---|---|---|
| NoSQL service | $0.50 / 1M Read Units | $1.50 / 1M Write Units | $0.35 / GB/month | Fully managed NoSQL database providing low latency at any scale |
| IndexedDB | $0.00 / 1M Read Units | $0.00 / 1M Write Units | $0.00 / GB/month | In-browser database engine |
| NoSQL service | $1.50 / 1M Read Units | $5.00 / 1M Write Units | $1.00 / GB/month | Distributed search and analytics engine |