Choose the AI models your team needs.
Choose from 240 documented AI model families and more than 44,000 documented deployment sizes, including more than 100 Meta AI model options — with guided server sizing for your workload.
Jet Models provide access to large and specialized AI models without requiring the customer to own the equivalent GPU hardware.
The customer still uses the familiar Terminal.Glass interface on one private Linux server, while model computation is handled through the supported cloud-model service.
General and Multimodal
Gemma 3 Cloud
Google Gemma 3 multimodal cloud variants for general chat, vision, and everyday business workflows.
Jet ModelGemma 4 Cloud
Google-backed multimodal model for general-purpose chat, images, and document understanding.
Jet ModelQwen 3.5 Cloud
Strong all-rounder for thinking, tools, coding, multilingual work, and business workflows.
Jet ModelGemini 3 Flash Preview Cloud
Speed-focused frontier preview model for fast responses and high-throughput chat.
Jet ModelKimi K2.5 Cloud
Native multimodal agentic model for chat, images, and mixed workloads.
Jet ModelMistral Large 3 Cloud
Production-grade multimodal MoE model for enterprise chat and analysis.
Coding and Software Engineering
Qwen3-Coder Cloud
Large coding model for repository work, multi-file edits, and software engineering agents.
Jet ModelMiniMax M2.5 Cloud
Productivity and coding model with long-context support for daily team work.
Jet ModelMiniMax M2.7 Cloud
Agentic coding model with long context for development and technical workflows.
Jet ModelGLM 4.7 Cloud
Technical coding and engineering model for software teams and IT workflows.
Jet ModelQwen3-Coder Next Cloud
Next-generation coding agent for autonomous development and repository work.
Jet ModelDevstral Small 2 Cloud
Codebase exploration and multi-file editing for everyday developer workflows.
Jet ModelRNJ-1 Cloud
Compact code and STEM model for focused technical and programming tasks.
Jet ModelKimi K2.6 Cloud
Long-horizon coding, design, and autonomous execution for complex projects.
Jet ModelDevstral 2 Cloud
Large coding agent model for repository-scale software engineering work.
Jet ModelMiniMax M3 Cloud
Frontier coding and multimodal agent model with very long context windows.
Jet ModelKimi K2.7 Code Cloud
Moonshot coding-focused agent for long-horizon development and multimodal agent work.
Reasoning and Agentic Work
GPT-OSS Cloud
Flagship cloud reasoning and agentic workloads for teams that need top-tier answers.
Jet ModelNemotron 3 Super Cloud
NVIDIA MoE model for efficient complex reasoning and multi-agent workflows.
Jet ModelGLM 5 Cloud
Large reasoning and agentic model for demanding analysis and engineering tasks.
Jet ModelGLM 5.1 Cloud
Agentic engineering model focused on coding, tools, and technical problem-solving.
Jet ModelDeepSeek V3.2 Cloud
Efficient reasoning and agent performance without local flagship-class hardware.
Jet ModelDeepSeek V3.1 Cloud
Hybrid reasoning model for long-context technical work, tools, and analysis.
Jet ModelNemotron 3 Nano Cloud
Efficient NVIDIA agentic model for capable workflows at lower cloud cost.
Jet ModelDeepSeek V4 Pro Cloud
Frontier MoE reasoning with long context for the most demanding workloads.
Jet ModelDeepSeek V4 Flash Cloud
Fast, efficient DeepSeek reasoning for responsive technical and business chat.
Jet ModelGLM 5.2 Cloud
Long-horizon engineering and agentic workflows for sustained technical projects.
Lightweight and Specialized
Browse documented private AI model families and deployment sizes through the NoCloudGPT model library — with Terminal.Glass guided sizing for your server.
These models are intended for customer-controlled hardware, private servers, or on-premises deployment.
Terminal.Glass provides the deployment and user experience. NoCloudGPT provides the deeper private-model reference library.
Not sure which models fit your workload?
Tell us what your team needs to accomplish and where you want the system hosted. We will recommend an appropriate starting set.
Ready to deploy?
Start with Sunrise Starter for $199 (2 Glass Licenses) or Sunrise Business for $399 (6 licenses). One license powers one active deployment.