CLI reference

2026-08-31

tempo9 (--gguf <.gguf> | --ollama <model[:tag]>) [options]
tempo9 --list-ollama

Choosing a model

FlagDefaultMeaning
--gguf <path>Model file. A bare .gguf path as the only argument works too.
--ollama <model[:tag]>A model Ollama already pulled. A bare name means :latest. See Ollama models.
--list-ollamaPrint what Ollama has pulled, then exit. Exits 1 if the store is empty or absent.
--graph <path>built from the GGUFPrecomputed graph. Rarely needed — the engine builds and caches one.
--name <string>the file's basename, or the Ollama nameWhat /v1/models and /api/tags report.

Serving

FlagDefaultMeaning
--port <n>11435Loopback port.
--max-length <n>32768Context length the engine is built for.
--max-batch <n>16Concurrent requests the engine schedules. 1 makes the server serial end to end.
--speculation-k <n>0MTP speculation depth (0 = off). The engine varies the effective k downward per request.
--tower <dir>Core ML vision tower. Required for image input.

There is no --token flag: nothing is authenticated. See Limits.

Environment

VariableEffect
OLLAMA_MODELSWhere to look for Ollama's store. Defaults to ~/.ollama/models.
AS_KV_OFFLOAD0 disables the on-by-default KV disk cache. See Limits.
AS_KV_OFFLOAD_GBDisk cap for that cache, default 16.
AS_KV_OFFLOAD_PATHIts location, default ~/Library/Caches/tempo9/kv.