Available Models

Arvae provides access to a wide range of AI models from leading providers, all available with a 15% discount compared to direct provider pricing.

This page lists all the models available through the Arvae AI API, along with their specifications and pricing information.

Hanooman Models

Hanooman Base Pro
Model ID
hanooman/hanooman-everest
Description
Specialized model with advanced capabilities
Input Price
Actual$2.00
Discounted$1.70
Output Price
Actual$9.00
Discounted$7.65
Hanooman Healthcare Expert
Model ID
hanooman/hanooman-everest-healthcare
Description
Specialized in medical knowledge, healthcare protocols, and clinical information. Only answers healthcare-related questions.
Input Price
Actual$3.00
Discounted$2.55
Output Price
Actual$12.00
Discounted$10.20
Hanooman Finance Expert
Model ID
hanooman/hanooman-everest-finance
Description
Expert in finance, economics, markets, and investment strategies. Only answers finance-related questions.
Input Price
Actual$3.00
Discounted$2.55
Output Price
Actual$12.00
Discounted$10.20
Hanooman Code Expert
Model ID
hanooman/hanooman-everest-code
Description
Proficient in software development, coding patterns, and technical problem-solving. Only answers programming-related questions.
Input Price
Actual$3.00
Discounted$2.55
Output Price
Actual$12.00
Discounted$10.20
Hanooman Conversation Expert
Model ID
hanooman/hanooman-everest-conversation
Description
Optimized for engaging, nuanced conversations with enhanced understanding of context. Focuses on conversation quality.
Input Price
Actual$3.00
Discounted$2.55
Output Price
Actual$12.00
Discounted$10.20

*Prices shown per 1M tokens. Green numbers show discounted rates (15% off standard pricing).

OpenAI Models

OpenAI O1
Model ID
openai/o1
Description
Most powerful OpenAI model
Input Price
Actual$15.00
Discounted$12.75
Output Price
Actual$60.00
Discounted$51.00
OpenAI O1 Mini
Model ID
openai/o1-mini
Description
Smaller version of O1
Input Price
Actual$1.10
Discounted$0.94
Output Price
Actual$4.40
Discounted$3.74
OpenAI GPT-4o
Model ID
openai/chatgpt-4o-latest
Description
Latest multimodal model with advanced capabilities
Input Price
Actual$5.00
Discounted$4.25
Output Price
Actual$15.00
Discounted$12.75
OpenAI GPT-4o Mini
Model ID
openai/gpt-4o-mini
Description
Smaller version of GPT-4o
Input Price
Actual$0.15
Discounted$0.13
Output Price
Actual$0.60
Discounted$0.51
OpenAI GPT-4.1
Model ID
openai/gpt-4.1
Description
Enhanced reasoning capabilities
Input Price
Actual$2.00
Discounted$1.70
Output Price
Actual$8.00
Discounted$6.80
OpenAI GPT-4.1 Mini
Model ID
openai/gpt-4.1-mini
Description
Smaller version of GPT-4.1
Input Price
Actual$0.40
Discounted$0.34
Output Price
Actual$1.60
Discounted$1.36
OpenAI GPT-4.1 Nano
Model ID
openai/gpt-4.1-nano
Description
Most compact version of GPT-4.1
Input Price
Actual$0.10
Discounted$0.09
Output Price
Actual$0.40
Discounted$0.34

*Prices shown per 1M tokens. Green numbers show discounted rates (15% off standard pricing).

Anthropic Models

Anthropic Claude 3.7 Sonnet
Model ID
anthropic/claude-3.7-sonnet
Description
Latest Claude model
Input Price
Actual$3.00
Discounted$2.55
Output Price
Actual$15.00
Discounted$12.75
Anthropic Claude 3.5 Haiku
Model ID
anthropic/claude-3.5-haiku
Description
Fast and cost-effective Claude model
Input Price
Actual$0.80
Discounted$0.68
Output Price
Actual$4.00
Discounted$3.40
Anthropic Claude 3.5 Sonnet
Model ID
anthropic/claude-3.5-sonnet
Description
Balance of intelligence and speed
Input Price
Actual$3.00
Discounted$2.55
Output Price
Actual$15.00
Discounted$12.75

*Prices shown per 1M tokens. Green numbers show discounted rates (15% off standard pricing).

Google Models

Google Gemini 2.0 Flash
Model ID
google/gemini-2.0-flash
Description
Fast and efficient Gemini model
Input Price
Actual$0.10
Discounted$0.09
Output Price
Actual$0.40
Discounted$0.34
Google Gemini 2.5 Pro Preview
Model ID
google/gemini-2.5-pro-preview-03-25
Description
Latest Gemini with advanced capabilities
Input Price
Actual$1.25
Discounted$1.06
Output Price
Actual$10.00
Discounted$8.50
Google Gemma 3 4B
Model ID
google/gemma-3-4b-it
Description
Compact, efficient model
Input Price
Actual$0.02
Discounted$0.02
Output Price
Actual$0.04
Discounted$0.03

*Prices shown per 1M tokens. Green numbers show discounted rates (15% off standard pricing).

Meta-llama Models

Meta Llama 3.3 70B
Model ID
meta-llama/llama-3.3-70b
Description
Largest Llama model
Input Price
Actual$0.10
Discounted$0.09
Output Price
Actual$0.25
Discounted$0.21
Meta Llama 3.1 405B
Model ID
meta-llama/llama-3.1-405b-instruct
Description
Massive Llama model with extensive capabilities
Input Price
Actual$0.80
Discounted$0.68
Output Price
Actual$0.80
Discounted$0.68

*Prices shown per 1M tokens. Green numbers show discounted rates (15% off standard pricing).

X-ai Models

xAI Grok-3 Beta
Model ID
x-ai/grok-3-beta
Description
Advanced reasoning and capabilities
Input Price
Actual$3.00
Discounted$2.55
Output Price
Actual$15.00
Discounted$12.75
xAI Grok-3 Mini Beta
Model ID
x-ai/grok-3-mini-beta
Description
Compact version with strong performance
Input Price
Actual$0.30
Discounted$0.26
Output Price
Actual$0.50
Discounted$0.43

*Prices shown per 1M tokens. Green numbers show discounted rates (15% off standard pricing).

Deepseek Models

DeepSeek Chat v3
Model ID
deepseek/deepseek-chat-v3
Description
Powerful conversational model
Input Price
Actual$0.27
Discounted$0.23
Output Price
Actual$1.10
Discounted$0.94
DeepSeek R1
Model ID
deepseek/deepseek-r1
Description
Advanced reasoning and capabilities
Input Price
Actual$0.54
Discounted$0.46
Output Price
Actual$2.18
Discounted$1.85

*Prices shown per 1M tokens. Green numbers show discounted rates (15% off standard pricing).

Hanooman Models

Using Hanooman Models

Hanooman models are accessed through the V2 endpoint. Make sure to use https://arvae.ai/api/v2/chat/completions when working with Hanooman models.

Hanooman models offer specialized capabilities designed for specific use cases. The expert models are domain-specific and will only respond to questions within their area of expertise.

Expert Models

Domain Restrictions

Expert models are designed to only answer questions within their specific domain of expertise. If you ask about topics outside their domain, they will politely decline and suggest consulting a different expert or general assistant.

🏥 Healthcare Expert

hanooman/hanooman-everest-healthcare

Specialized in medical knowledge, healthcare protocols, clinical guidelines, medical research, anatomy, physiology, pharmacology, and healthcare systems.

Domain: Healthcare, Medicine, Wellness, Medical Research

💰 Finance Expert

hanooman/hanooman-everest-finance

Expert in finance, economics, investments, markets, banking, accounting, financial planning, and business strategy.

Domain: Finance, Economics, Investments, Banking, Accounting

💻 Code Expert

hanooman/hanooman-everest-code

Proficient in software development, programming languages, algorithms, system design, debugging, and software engineering best practices.

Domain: Programming, Software Development, Computer Science

💬 Conversation Expert

hanooman/hanooman-everest-conversation

Optimized for engaging, nuanced conversations with enhanced understanding of context, communication nuance, and conversation flow.

Domain: Natural Dialogue, Conversation Quality, Communication

How to Choose an Expert Model

  • Healthcare Expert: Use for medical questions, health information, clinical guidance
  • Finance Expert: Use for financial analysis, investment advice, economic questions
  • Code Expert: Use for programming help, code review, technical problem-solving
  • Conversation Expert: Use for general conversation, creative writing, communication tasks
  • Base Model: Use for general-purpose tasks that don't require domain expertise

Using Models in Your Requests

To use a specific model, include the model ID in the model parameter of your API request. For example:

{ "model": "hanooman/hanooman-everest-healthcare", "messages": [ {"role": "user", "content": "What are the symptoms of diabetes?"} ] }

Remember to use the V2 endpoint for Hanooman models and the V1 endpoint for all other models. Expert models will only respond to questions within their domain of expertise.