Skip to content

Configuration ​

Every orchestrator setting is an environment variable and a flag of the same name: DPPAN_JOB_MAX_QUEUED=128 or --job-max-queued 128 (a flag wins). Node and download settings are environment variables on the machine they apply to. The dashboard lists this deployment's limits under Settings → Limits.

Server ​

VariableDefault
PORTrequiredThe orchestrator's REST API port
GATEWAY_PORTPORT + 1000The port nodes dial in on, so they need no open ports of their own
DPPAN_ADMIN_TOKENnoneTurns on admin endpoints and the dashboard's Admin tab
DPPAN_MODELS_CONFIGconfig/models.toml (the current directory, else next to the binary)The model registry
DPPAN_GGUF_<MODEL>noneThe orchestrator's file for a model, e.g. DPPAN_GGUF_LLAMA3_3_70B for llama3.3:70b: see Sharded download
DPPAN_MODELS_DIR~/.dppan/modelsWhere pull installs and the orchestrator looks, in bundles/
LOG_LEVELinfotrace, debug, info, warn or error
HEALTH_INTERVAL_SECS10How often the orchestrator checks its nodes
INCLAVATE_NO_BANNERunset1 hides the startup banner, as --no-banner does

Chat ​

VariableDefault
MAX_TOKENS512Most tokens a reply may have. A reasoning model spends its thinking from the same budget, so raise it if replies come back empty
MAX_CONTEXT_TOKENS2048How much of a conversation is kept; past it, the oldest turns are dropped. 0 keeps all of it. Nodes that set DPPAN_CTX need it at least this high: raise the two together
DPPAN_RANK_TPS_WEIGHT1.0Multiplier on throughput (the numerator). Raise it to favour faster nodes more aggressively
DPPAN_RANK_LATENCY_WEIGHT1.0Multiplier on latency (the denominator). Set below 1.0 to de-emphasise latency

Jobs ​

The queue that speech, music, images, video and understanding run through: Generation jobs.

VariableDefault
DPPAN_JOB_MAX_QUEUED64Waiting jobs allowed; running jobs do not count
DPPAN_INTERACTIVE_LANEonoff makes speech, voice cloning and transcription wait for a running music, image or video job on the same machine
DPPAN_JOB_DEADLINE_MINUTES10Deadline, from submission, for jobs without a time limit of their own, such as qwen3tts and orpheus speech; 0 for none. Image and video jobs use DPPAN_VISUAL_DEADLINE_MINUTES. A request's deadline_seconds replaces either, up to 24 h
DPPAN_CLEANUP_INTERVAL_SECS300How often expired artifacts and jobs are removed; 0 never
DPPAN_WORKER_IDkept in DPPAN_ARTIFACT_ROOTSet it only when that directory does not survive a restart (a container with no volume), or unfinished jobs stay running

Artifacts ​

Uploads and results: Artifacts.

VariableDefault
DPPAN_ARTIFACT_ROOT~/.dppan/artifactsWhere artifacts are stored
DPPAN_ARTIFACT_MAX_BYTES0 (unlimited)Storage allowed per owner, in bytes
DPPAN_ARTIFACT_MAX_OBJECTS0 (unlimited)Artifacts allowed per owner
DPPAN_ARTIFACT_TTL_HOURS0 (keep forever)How long new artifacts are kept

Speech ​

More on each: Speech and voice › Configuration.

VariableDefault
DPPAN_SPEECH_VOICESunsetNamed clips: name=/path/clip.wav, comma-separated
DPPAN_SPEECH_TEMPERATURE0.9Default sampling temperature
DPPAN_SPEECH_TOP_K50<= 0 disables it
DPPAN_SPEECH_TOP_P1.0>= 1.0 disables it
DPPAN_SPEECH_REPETITION_PENALTY1.051.0 disables it
DPPAN_SPEECH_MAX_ATTEMPTS1Retries for a model that does not stop: see When a model will not stop
DPPAN_SPEECH_MAX_FRAMES512Length ceiling per piece (~41 s on qwen3tts, ~44 s on orpheus): see Long text
DPPAN_SPEECH_SYNC_TIMEOUT_SECS120How long /v1/audio/speech waits before returning a job ID to poll
DPPAN_SPEECH_MEMORY_BUDGET_MBhalf of RAMMemory all loaded speech, music, image and video models may hold together
DPPAN_SPEECH_MODELunsetBackbone GGUF. Overrides discovery
DPPAN_SPEECH_MMPROJunsetIts decoder GGUF, required alongside it
DPPAN_SPEECH_MODEL_IDqwen3-ttsThe id requests use for that pair

Music ​

More on each: Music › Configuration.

VariableDefault
DPPAN_MUSIC_BUNDLEunsetModel directories, comma-separated. Unset uses ~/.dppan/models/bundles
DPPAN_MUSIC_MAX_FRAMES250 (10 s)Longest song, in frames at 25 Hz
DPPAN_MUSIC_STEPS0 (the model's own)Default solver steps
DPPAN_MUSIC_THREADS0 (the machine's)CPU threads for music
DPPAN_MUSIC_SYNC_TIMEOUT_SECS900How long /v1/audio/music waits before returning a job ID to poll

Images and video ​

More on each: Image and video settings › Configuration.

VariableDefault
DPPAN_VISUAL_MAX_VRAM_GIB0 (what is free)GPU memory per image or video model, in GiB; a negative value leaves that much free. Unset, a model leaves room for the chat of text nodes on the same machine
DPPAN_VISUAL_NODESoffon accepts nodes' image and video models: see Image and video nodes
DPPAN_VISUAL_PREFER_NODESonJobs that name no place try nodes serving only image and video models before this machine. off tries this machine first
DPPAN_VISUAL_MAX_FRAMES481 (20 s at 24 fps)Longest video a request may ask for, 1 to 1024. A model trained on shorter clips stops at its own length
DPPAN_VISUAL_DEADLINE_MINUTES240 (4 h)How long an image job may run, and a video for each part it is made in, 1 to 1440
DPPAN_VISUAL_IDLE_MINUTES10How long an image or video model stays loaded with no job, 1 to 1440. Then its process stops and the next job loads it again
DPPAN_IMAGES_SYNC_TIMEOUT_SECS900How long /v1/images/generations and /v1/images/edits wait before returning a job ID to poll

Nodes ​

Set on the machine running dppan join. Its flags are in the CLI reference.

VariableDefault
ORCHESTRATOR_URLnoneThe orchestrator, as --orchestrator
DPPAN_MODELllama3.2:1bThe model, as --model
GGUF_PATHnoneA model file already downloaded, as --gguf
GATEWAY_URLthe orchestrator's host, port + 1000Where to dial the orchestrator's gateway
DPPAN_CTXthe model's ownThe context each chat session is given memory for. Lower it on a machine short of memory, but not below the orchestrator's MAX_CONTEXT_TOKENS
DPPAN_VISUAL_MAX_VRAM_GIB0 (what is free)On a node offering image and video models: GPU memory per model, as on the orchestrator

Downloads ​

VariableDefault
HF_TOKENunsetHugging Face token for gated repositories

Free to run · Proprietary