Ollama, LM Studio, Jan.ai and GPT4All are runners — install a runner and launch any model in it. The system prompt from an AILoft agent works in any of these tools.

MetaLlama 4, Llama 3.3, Llama 3.2, Llama 3.1, Llama 2

MicrosoftPhi-4, Phi-4 Mini, Phi-3, WizardLM2

GoogleGemma 3, Gemma 3n, Gemma 2, CodeGemma

AlibabaQwen3.5, Qwen3, Qwen2.5, Qwen2.5-Coder, QwQ

MistralMistral Large, Mistral Small, Mixtral, Magistral, Codestral, Devstral

DeepSeekDeepSeek-R1, DeepSeek-V3, DeepSeek-Coder-V2, DeepCoder

IBMGranite 4, Granite 3.3, Granite Code

OpenAIGPT-OSS

CommunityDolphin 3, Hermes3, Cogito, OpenThinker

OtherLLaVA, CodeLlama, StarCoder2, TinyLlama, Falcon3, SmolLM2, OLMo2, Yi, Orca Mini, MiniCPM-V, Moondream, LFM2, Command-R, Zephyr

How Installation Works

Choose Your Agent

Browse the marketplace and pick an agent that fits your needs.

Select a Backend

Connect your preferred AI provider — ChatGPT, Claude, Gemini, or others.

Deploy Instantly

Install with one click. Run the agent in your LLM backend.

Browse Agents →