Skip to content

Providers & models

agent-mem has no API key and no model of its own. It starts the CLI you already use, under your login, with no tools, no MCP servers, no settings and no saved session, in an empty temporary folder.

Claude Code requests to claude -p with haiku, Codex requests to codex exec with gpt-6-luna, Copilot requests to copilot -s auto, OpenCode requests to opencode run with the free big-pickle model; Ollama only when added.
Open in Excalidraw
Request came fromModelCommand
Claude CodeClaude Haikuclaude -p --restricted --setting-sources "" --tools "" --strict-mcp-config --no-session-persistence --model haiku
CodexGPT-6-Luna, low effortcodex exec --ephemeral --ignore-user-config --disable hooks --disable memories -s read-only -m gpt-6-luna -c model_reasoning_effort=low
Copilot CLI / appauto, efficiency tier (resolves to GPT-6-Luna)copilot -s --no-custom-instructions --disable-builtin-mcps --model auto --auto-tier efficiency
OpenCodeBig Pickle, freeopencode run --standalone --agent plan --model opencode/big-pickle

A request never leaves its vendor. A Claude Code request is not sent to OpenAI, a Codex request not to Anthropic. Each uses the cheapest model of its CLI. With only one CLI installed, its requests get facts and the others are saved word for word.

config.json
{
"observe": {
"claude_model": "sonnet",
"codex_model": "gpt-6",
"copilot_model": "auto"
}
}

If your plan does not offer a model, or the CLI is too old for the flag, the call is tried once more with the CLI’s own default. agent-mem doctor and the session panel name the model in use: Facts written by haiku (claude).

config.json
{ "observe": { "providers": ["opencode", "claude", "codex", "copilot", "ollama"], "ollama_model": "qwen3.5:4b-mlx" } }

Ollama is off by default because local models load the machine. Added, it is the fallback for any request whose own CLI is missing, paused or fails. It is not counted against the daily limit.

config.json
{ "observe": { "cross_vendor": true } }

Then the first installed CLI in observe.providers takes every request, whatever agent it came from. Off by default, because it sends one vendor’s conversation to another.

Facts use the subscription or plan you already have, through the CLI’s own login.

Per batch of up to 5 requestsCalls
facts1
titles, session summary, lessons1
per project and day1 review, up to 3 graph batches of 40 memories

observe.daily_limit (300) caps model calls per day. agent-mem observe --status shows today’s calls per provider. A usage limit pauses that provider for observe.pause_minutes (60); requests wait rather than being lost.