Skip to content

Supported Providers

LongerAgent supports 10 providers across cloud APIs and local inference servers. Use longeragent init to configure any combination of them.

Provider Table

ProviderModelsAuth
AnthropicClaude Haiku 4.5, Opus 4.6, Sonnet 4.6 (+ 1M context variants)ANTHROPIC_API_KEY
OpenAIGPT-5.2, GPT-5.2 Codex, GPT-5.3 Codex, GPT-5.4OPENAI_API_KEY or OAuth
Kimi / MoonshotKimi K2.5, K2 Instruct (Global, China, Coding Plan)Managed slots (LONGERAGENT_KIMI_*); imports MOONSHOT_API_KEY / KIMI_* during setup
MiniMaxM2.1, M2.5 (Global, China)Managed slots (LONGERAGENT_MINIMAX_*); imports MINIMAX_* during setup
GLM / ZhipuGLM-5, GLM-4.7 (Global, China, Coding Plan)Managed slots (LONGERAGENT_GLM_*); imports GLM_* during setup
OllamaAny local Ollama model (dynamic discovery)--
oMLXAny local MLX model (dynamic discovery)--
LM StudioAny local GGUF model (dynamic discovery)--
OpenRouterCurated presets for Claude, GPT, Kimi, MiniMax, GLM, plus any custom modelOPENROUTER_API_KEY

Cloud vs. Local

Cloud providers (Anthropic, OpenAI, Kimi, MiniMax, GLM, OpenRouter) require an API key. The init wizard prompts you for the key and stores it in ~/.longeragent/.env. GLM, Kimi, and MiniMax use LongerAgent-managed internal slots there instead of relying on runtime provider-to-env mappings.

Local providers (Ollama, oMLX, LM Studio) connect to a server running on your machine. No API key is needed. During longeragent init, the wizard queries the server's /v1/models endpoint to discover available models.

Switching at Runtime

Use the /model command during a session to switch between any configured model. For GLM, Kimi, and MiniMax, selecting a model with a missing key can prompt you to import or paste the key on the spot. Inline /model ... key=... syntax is not supported. You can also adjust reasoning depth with /thinking.

See Model Switching for details.

Setup Guides

Released under the MIT License.