Skip to content

AI Provider

The AI Provider tab lets you choose which AI service powers your chatbot and configure related options.

Navigate to: Inqyra > Configuration > AI Provider tab

The AI Provider tab showing provider selection, model dropdown, and query rewriting options The AI Provider tab with provider selection and model configuration

Choosing a Provider

Select one of the four supported AI providers:

Provider Comparison

Provider Strengths API Key Source
Claude (Anthropic) Excellent reasoning, natural tone, strong instruction following console.anthropic.com
OpenAI Wide model range, well-known, good general performance platform.openai.com
Google Gemini Cost-effective, fast, good multilingual support aistudio.google.com
Mistral European provider, competitive pricing, fast inference console.mistral.ai

Note

You need to enter an API key for your chosen provider on the Settings page before it becomes available here. Providers without a configured API key are shown as unavailable.

Model Selection

After selecting a provider, choose a specific model from the dropdown. Each model shows its pricing per 1 million tokens.

Available Models

Model Input (per 1M tokens) Output (per 1M tokens)
Claude Sonnet 4 $3.00 $15.00
Claude Opus 4 $15.00 $75.00
Claude 3.5 Sonnet $3.00 $15.00
Claude 3.5 Haiku $0.80 $4.00
Claude 3 Opus $15.00 $75.00
Claude 3 Sonnet $3.00 $15.00
Claude 3 Haiku $0.25 $1.25
Model Input (per 1M tokens) Output (per 1M tokens)
GPT-4o $2.50 $10.00
GPT-4o Mini $0.15 $0.60
GPT-4 Turbo $10.00 $30.00
GPT-4 $30.00 $60.00
GPT-3.5 Turbo $0.50 $1.50
o1 $15.00 $60.00
o1 Mini $3.00 $12.00
Model Input (per 1M tokens) Output (per 1M tokens)
Gemini 2.5 Flash $0.15 $0.60
Gemini 2.5 Flash Lite $0.075 $0.30
Gemini 2.5 Pro $1.25 $5.00
Gemini 2.0 Flash $0.10 $0.40
Gemini 2.0 Flash Lite $0.075 $0.30
Model Input (per 1M tokens) Output (per 1M tokens)
Mistral Large $2.00 $6.00
Mistral Small $0.20 $0.60
Codestral $0.30 $0.90
Ministral 8B $0.10 $0.10
Ministral 3B $0.04 $0.04
Pixtral Large $2.00 $6.00
Open Mistral Nemo $0.15 $0.15

Click Refresh Models to update the model list if new models have been added by the provider.

Choosing a Model

For most use cases, a mid-range model offers the best balance of quality and cost. Consider:

  • Best quality: Claude Opus 4, GPT-4o, Gemini 2.5 Pro, Mistral Large
  • Best value: Claude 3.5 Haiku, GPT-4o Mini, Gemini 2.5 Flash Lite, Ministral 8B
  • Balanced: Claude Sonnet 4, GPT-4o, Gemini 2.5 Flash, Mistral Small

Query Rewriting

Query rewriting is an optional feature that improves search accuracy by rephrasing visitor questions before searching your knowledge base.

How It Works

When a visitor asks a vague or conversational question like "what about returns?", the query rewriter transforms it into a more specific search query like "What is the return and refund policy?". This helps the search engine find more relevant content.

Configuration

  1. Check Enable Query Rewriting
  2. Select a Rewrite Model — use a fast, inexpensive model since this runs on every message:
    • Claude 3.5 Haiku
    • GPT-4o Mini
    • Gemini 2.5 Flash Lite
    • Ministral 8B

Cost Impact

Query rewriting adds a small additional cost per message since it makes an extra API call. Using a lightweight model keeps this cost minimal (typically less than $0.001 per query).