Models / Higgsfield / Reference to video
Higgsfield reference to video prompting guide
Last checked 2026-10-11Higgsfield (host platform) · Host: Kling 3.0/O3/2.6, Seedance 2.0/2.5, Wan 2.6/2.7/3.0, MiniMax H3/Hailuo 2.3, LTX-2.5, Grok Imagine 1.5, PixVerse V6, Happy Horse; own: Cinema Studio 4.0, Genjutsu
How to prompt Higgsfield in reference-to-video mode. Every point links its source; community tips are marked. Your agent gets the same guide through the Atlas MCP (get_guide).
Prompt structure
ref tokens→subject with ref→action→camera→audio
- Higgsfield hosts 22 video model families (Seedance, Kling, Wan, MiniMax, LTX, Grok, PixVerse, Happy Horse) plus its own Cinema Studio 4.0 and Genjutsu; each workflow has its own schema. source ↗
What the text should describe
- Cinema Studio 4.0 tokens: <<<image_1>>>, <<<video_1>>>, <<<audio_1>>> (one-based, must match an attached item of that type); up to 30 images, 10 videos, 10 audio, 50 total. source ↗
- Seedance 2.5 reference-to-video: at least one non-empty image_urls/video_urls/audio_urls; audio-only is allowed; video and audio each normalized to ≤30 s total. source ↗
- MiniMax H3: resolution fixed '2K'; audio refs need an image or video ref; refs may be cropped/trimmed to a 15 s budget. source ↗
Duration and limits
- Kling on Higgsfield: chain scenes with the final frame as next start frame; add characters as Elements. source ↗
- Kling on Higgsfield: top-level prompt ≤2,500 chars (truncated), multi_prompt shot ≤512 chars, 1–6 shots. source ↗
Negative prompt
Supported via .
Real prompts, credited
A slow cinematic camera move around the subject in warm evening light.
Next
All Higgsfield modesHiggsfield model, prices & shotsFrom $0.0578/s · pricesSeedance reference to videoKling reference to videoVeo reference to videoHailuo (MiniMax) reference to video
Sources
- docs.higgsfield.ai/docs/models/cinema-studio-4/generate
- docs.higgsfield.ai/docs/models/kling-3/pro-image-to-video
- docs.higgsfield.ai/docs/models/kling-o3
- docs.higgsfield.ai/docs/models/minimax-h3/reference-to-video
- docs.higgsfield.ai/docs/models/seedance-2-5/reference-to-video
- docs.higgsfield.ai/docs/models/wan-3/image-to-video
- higgsfield.ai/creator-hub/help-center/ai-models/how-do-i-use-kling