Models.
NGen 4 Models.
Massive scale non-reasoning models for conversational AI
Conversational AI. 100B+ params. High performance, low latency.
Frontier non-reasoning model. Massive scale, exceptional capabilities.
NGen 3.5 Series Models.
Advanced conversational models with enhanced capabilities
NGen 3.5 Max
Maximum performance and capability for complex reasoning.
NGen 3.5 Pro
Professional-grade conversational AI for advanced workflows.
NGen 3.5 Lite
Efficient and fast conversational model for real-time use.
NGen 3.1 Models.
Next-generation multimodal models with breakthrough performance
NGen3.1-Pro Multimodal
Handles image and text inputs with breakthrough performance.
Omni Models.
Real-time multimodal and audio processing capabilities
NGen-Realtime-Omni
Text, audio, and image in real-time.
NGen-Realtime-Omni-Audio
Real-time audio processing and understanding.
Audio Models.
Advanced audio processing and text-to-speech capabilities
AudioTextSpeech-Pro
Professional-grade text-to-speech synthesis with natural prosody.
AudioTextSpeech-1
Basic text-to-speech for simple applications and notifications.
Image Models.
Advanced image generation capabilities
IGen-1 Nano
512×512, fast and efficient for rapid prototyping.
IGen-1
Up to 4K resolution with advanced editing and prompt adherence.
NGen 3.9 Models.
Latest generation conversational AI models
NGen 3.9 Lite
Fast & efficient for basic chat and low-latency interaction.
NGen 3.9 Pro
Strong contextual and logical understanding for professional use.
NGen 3.9 Max
Stable enterprise version for large-scale production workloads.
NGen 3 Models.
NGen 3 10B
Frontier Model for advanced reasoning and large-scale applications.
NGen 3 7B
Foundational Model balancing power and speed.
NGen 3 3B
Accessible AI for general-purpose chat and assistance.
NGen 3 140M
Compact model for efficient on-device processing.
NGen 3 140M V2
Enhanced compact model with improved instruction following.
NGen 3 90M
Smart micro-model for simple text transformations.
NGen 3 15M
Micro-scale intelligence for highly constrained environments.
NGen 3 15M Embed
Specialized vector search and embedding model.
NGen 3 140M Instruct
Task-focused model optimized for direct instructions.
NGen3-7B-0625 Chat
High-efficiency conversational model with modern tuning.
NGen3-1B Chat
Compact 1B parameter model for high-speed edge deployment.
NGen3-1B-it Chat
Instruction-tuned variant for superior task performance.
NGen 2 Models.
NGen 2 170M
Gen-2 Core for lightweight language tasks.
NGen 2 30M
Compact Core for on-device processing.
NGen 2 15M
Micro-Scale for extremely constrained environments.
Stellar - 2.
Stellar Large
Stellar Scale for enterprise-grade reasoning.
Stellar Mini
Compact Power with smart reasoning capabilities.
Stellar Nano
Ultra-Compact edge-native intelligence.
Special Tool Models.
Specialized models for search, research, code, and agent coordination
WebSearch-1
Real-time web search with source citations.
Scientist-1
Specialized for scientific research and analysis.
Code-1
Supports 30+ programming languages with advanced debugging.
MultiAgent-1
Coordinates multiple AI agents for complex tasks.
AGent*
Autonomous strategic agent for complex decision making.
NGen 1 Models.
Pioneer foundation models that started it all
NGen 1
Pioneer foundation model that started it all.