Configuration#

OpenMontage reads global settings from config.yaml and provider credentials from .env. These files control LLM behavior, spending limits, state persistence, output defaults, and platform-specific render profiles. All config.yaml values are validated by Pydantic models in lib/config_model.py.

config.yaml#

Place config.yaml in the project root. The file defines five top-level sections.

llm:
  provider: anthropic
  temperature: 0.7
  max_tokens: 4096

budget:
  mode: warn
  total_usd: 10.00
  reserve_pct: 0.10
  single_action_approval_usd: 0.50

checkpoint:
  policy: guided
  storage_dir: pipeline

output:
  default_format: mp4
  default_codec: libx264
  default_audio_codec: aac
  default_resolution: 1920x1080
  default_fps: 30
  default_crf: 23

paths:
  pipeline_dir: pipeline
  library_dir: library
  styles_dir: styles
  skills_dir: skills
  output_dir: output

llm sets the coding assistant provider and generation parameters.
budget controls spending mode (observe, warn, or cap), total limit, reserve holdback, and approval threshold.
checkpoint selects the default policy (guided, manual_all, or auto_noncreative) and storage location.
output supplies default render parameters used when no platform profile is selected.
paths overrides directory locations for pipeline state, libraries, styles, skills, and final output.

.env File#

Copy the example and add only the keys you need:

cp .env.example .env

Variable	Used By	Purpose
`ELEVENLABS_API_KEY`	elevenlabs_tts, music_gen	TTS, music, sound effects
`OPENAI_API_KEY`	openai_tts, openai_image	TTS fallback, DALL-E 3
`XAI_API_KEY`	grok_image, grok_video	Grok image editing/generation and video
`FAL_KEY`	flux_image, kling_video, veo_video, minimax_video, recraft_image	fal.ai hosted models
`HEYGEN_API_KEY`	heygen_video	Multi-provider video gateway
`PEXELS_API_KEY`	pexels_image, pexels_video	Free stock media
`PIXABAY_API_KEY`	pixabay_image, pixabay_video	Free stock media
`GOOGLE_API_KEY`	google_imagen, google_tts	Imagen images and Cloud TTS
`RUNWAY_API_KEY`	runway_video	Runway Gen-4 video
`HIGGSFIELD_API_KEY` + `HIGGSFIELD_API_SECRET`	higgsfield_video	Higgsfield multi-model video
`SUNO_API_KEY`	suno_music	Full song generation
`MODAL_LTX2_ENDPOINT_URL`	ltx_video_modal	Self-hosted LTX-2
`VIDEO_GEN_LOCAL_ENABLED`	local video tools	Enable local GPU generation
`VIDEO_GEN_LOCAL_MODEL`	wan, hunyuan, ltx, cogvideo	Select local model
`DOUBAO_SPEECH_API_KEY` / `DOUBAO_SPEECH_VOICE_TYPE`	doubao_tts	Mandarin TTS
`HF_TOKEN`	analysis tools	Optional speaker diarization
`UNSPLASH_ACCESS_KEY`	stock tools	Free stock images

See Provider Guide for setup steps and pricing per key. See Configuring Providers for first-time setup.

Media Profiles#

lib/media_profiles.py supplies platform-specific render configurations. Each profile defines resolution, aspect ratio, codec, audio codec, CRF, pixel format, max file size, max duration, and caption format. The ffmpeg_output_args(profile) helper produces the matching FFmpeg flags.

Profile	Resolution	Aspect	Notes
youtube_landscape	1920x1080	16:9	Standard YouTube
youtube_4k	3840x2160	16:9	4K YouTube
youtube_shorts	1080x1920	9:16	Max 60s
instagram_reels	1080x1920	9:16	Max 90s
instagram_feed	1080x1080	1:1	Square
tiktok	1080x1920	9:16	Vertical
linkedin	1920x1080	16:9	Landscape
cinematic	2560x1080	21:9	Ultrawide

Profiles are referenced from edit_decisions and passed to composition tools. See Media Profiles for the complete field list per profile.