Inflection 3 Pi
Description
Inflection 3 Pi powers Inflection's Pi chatbot, including backstory, emotional intelligence, productivity, and safety. It has access to recent news, and excels in scenarios like customer support and roleplay. The system is designed to mirror user communication styles, including emoji usage patterns.
Key Capabilities
- Emotional Intelligence: Designed for empathetic, natural conversations
- Customer Support: Excels at handling support scenarios
- Roleplay: Strong performance in roleplay interactions
- Communication Mirroring: Adapts to user's communication style including emoji patterns
- Current Events: Has access to recent news
Technical Specifications
| Specification |
Value |
| Context Window |
8,000 tokens |
| Context Length |
8,000 tokens |
| Max Completion Tokens |
1,024 tokens |
| Input Modalities |
Text |
| Output Modalities |
Text |
Pricing
| Type |
Cost per Million Tokens |
| Input Tokens |
$2.50 |
| Output Tokens |
$10.00 |
Supported Parameters
| Parameter |
Supported |
max_tokens |
Yes |
temperature |
Yes |
top_p |
Yes |
stop (sequences) |
Yes |
Model Identification
| Property |
Value |
| Name |
Inflection: Inflection 3 Pi |
| Short Name |
Inflection 3 Pi |
| Model ID |
inflection/inflection-3-pi |
| Creator/Author |
Inflection |
| Created |
October 11, 2024 |
| Model Group |
Other |
| Property |
Value |
| Provider Name |
Inflection |
| Provider Slug |
inflection |
| Adapter |
InflectionLegacyAdapter |
| Base URL |
https://layercake.pubwestus3.inf7ks8.com/external/api/inference |
The model supports tool choice specifications:
literal_none - No tool use
literal_auto - Automatic tool selection
literal_required - Tool use required
type_function - Function-type tools
Features
| Feature |
Status |
| Chat Completions |
Supported |
| BYOK (Bring Your Own Key) |
Enabled |
| Reasoning Support |
Not Supported |
Data Policy
Usage Notes
Best Use Cases
- Conversational AI: Natural, empathetic dialogue
- Customer Support: Handling user queries with emotional intelligence
- Roleplay Scenarios: Character-based interactions
- Personal Assistant: Productivity and general assistance
Limitations
- Relatively short context window (8,000 tokens)
- Limited max completion tokens (1,024)
- Text-only modality (no image/audio support)
- No reasoning/chain-of-thought capabilities
API Example
curl -X POST https://api.langmart.ai/v1/chat/completions \
-H "Authorization: Bearer $LANGMART_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "inflection/inflection-3-pi",
"messages": [
{"role": "user", "content": "Hello, how are you today?"}
],
"temperature": 0.7,
"max_tokens": 512
}'
Source