> ## Documentation Index
> Fetch the complete documentation index at: https://hedra.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Which Video Model Should I Use?

> A guide to choosing the right video model in Hedra for your project.

# Hedra Video Model Guide

There isn’t one “best” model. There is a best model for your goal.

If you’re unsure where to begin:

• Dialogue or presenters → Start with Hedra-native models\
• Cinematic visuals → Use text-to-video tiers\
• Copying motion → Use Motion / Reference models\
• Testing ideas → Use budget or fast tiers

***

# Hedra Native Models (Recommended Starting Point)

***

### [Hedra Avatar](https://www.hedra.com/app/video?model=hedra-avatar) *-7 credits per second*

**Category:** Audio-Driven Avatar\
**Audio Support:** Yes\
**Input Type:** Image + Audio

**Best For:**\
Talking heads, presenters, tutorials, explainers.

**Max Duration:** 10 minutes per clip

**Strengths:**\
• Reliable lip-sync\
• Stable performance delivery\
• Cost-efficient for longer dialogue\
• Longest single-clip duration of any model on Hedra

**Use When:**\
You need a host speaking clearly on camera, or you need long-form uncut video.

**Avoid When:**\
You’re building complex cinematic scenes.

***

### [Hedra Omnia](https://www.hedra.com/app/video?model=hedra-omnia) *-15 credits per second*

**Category:** Cinematic Dialogue\
**Audio Support:** Yes\
**Input Type:** Text + Audio

**Best For:**\
Short cinematic scenes with expressive dialogue and controlled camera movement.

**Strengths:**\
• Lifelike motion and emotion\
• Precision camera control\
• Strong for short-form ads

**Use When:**\
You want dialogue plus dynamic framing.

**Avoid When:**\
You need long monologues or large-scale action sequences.

***

# Premium Cinematic Models

***

### [Sora 2 Pro](https://www.hedra.com/app/video?model=sora-2-pro) *-70 credits per second*

**Category:** Premium Narrative\
**Audio Support:** Confirm\
**Input Type:** Text-to-video

**Best For:**\
Complex narrative scenes and high realism.

**Strengths:**\
• Strong scene coherence\
• Advanced world interaction

**Use When:**\
Rendering final, high-impact cinematic shots.

**Avoid When:**\
Testing prompts or simple dialogue.

***

### [Veo 3](https://www.hedra.com/app/video?model=veo-3) / [Veo 3.1](https://www.hedra.com/app/video?model=veo-3-1) *-55 credits per second*

**Category:** Premium Cinematic\
**Audio Support:** No\
**Input Type:** Text-to-video

**Best For:**\
Photorealistic environments and physics-heavy scenes.

**Strengths:**\
• High realism\
• Strong camera motion\
• Environmental detail

**Use When:**\
You need polished visual realism.

**Avoid When:**\
You need native speech performance.

***

### [Kling V3 Pro](https://www.hedra.com/app/video?model=kling-v3-pro) *-40 credits per second*

**Category:** Premium Cinematic\
**Audio Support:** Yes (Pro tier)\
**Input Type:** Text-to-video

**Best For:**\
High-resolution cinematic sequences with audio.

**Strengths:**\
• Strong visual fidelity\
• Extended generation capability

**Use When:**\
Producing final cinematic visuals with dialogue.

**Avoid When:**\
You are experimenting or testing cheaply.

***

# Balanced Cinematic Models

***

### [Kling O3 Pro](https://www.hedra.com/app/video?model=kling-o3-pro) *-24 credits per second*

**Category:** Cinematic Text-to-Video\
**Audio Support:** Yes\
**Input Type:** Text-to-video

**Best For:**\
Balanced cinematic scenes with audio support.

**Use When:**\
You want good quality without premium pricing.

**Avoid When:**\
You need maximum realism or budget testing.

***

### [Veo 3 Fast](https://www.hedra.com/app/video?model=veo-3-fast) / [Veo 3.1 Fast](https://www.hedra.com/app/video?model=veo-3-1-fast) *-20 credits per second*

**Category:** Fast Cinematic\
**Audio Support:** No\
**Input Type:** Text-to-video

**Best For:**\
Quick-turn cinematic visuals.

**Avoid When:**\
You need maximum fidelity.

***

# Budget & Fast Iteration

***

### [Kling V3 Standard](https://www.hedra.com/app/video?model=kling-v3-standard) -*9 credits per second*

**Category:** Balanced Cinematic\
**Audio Support:** No\
**Input Type:** Image-to-video (start frame only)

**Best For:**\
Animating a strong starting image into a short clip.

**Use When:**\
You already have a polished still frame.

**Avoid When:**\
You need dialogue or extended scene control.

***

# Motion Transfer & Reference Models

***

### [Kling O3 Reference](https://www.hedra.com/app/video?model=kling-o3-reference) -*24 credits per second*

**Category:** Reference Preservation\
**Audio Support:** Confirm\
**Input Type:** Reference Video

**Best For:**\
Preserving camera style and motion from an existing clip.

***

### [Kling O3 Edit](https://www.hedra.com/app/video?model=kling-o3-edit) / [O3 Standard Edit](https://www.hedra.com/app/video?model=kling-o3-standard-edit) *18-24 credits per second*

**Category:** Video Editing\
**Audio Support:** Confirm\
**Input Type:** Existing Video

**Best For:**\
Editing videos using natural language instructions.

***

### [Kling O1](https://www.hedra.com/app/video?model=kling-o1) -*20 credits per second*

**Category:** Start-End Frame Animation\
**Audio Support:** No\
**Input Type:** Start Frame + End Frame

**Best For:**\
Animating between two images.

***

### [Seedance 1.5 Pro](https://www.hedra.com/app/video?model=seedance-1-5-pro) -*8 credits per second*

**Category:** Motion + Audio\
**Audio Support:** Yes\
**Input Type:** Start/End Frame + Audio

**Best For:**\
Audio-driven motion between defined frames.

***

# Avatar Alternatives

***

### [Kling AI Avatar v2 Pro](https://www.hedra.com/app/video?model=kling-ai-avatar-v2-pro) -*16 credits per second*

**Category:** Avatar\
**Audio Support:** Yes\
**Input Type:** Image + Audio

**Best For:**\
Creating avatar videos from still images.

***

### [Kling AI Avatar v2 Standard](https://www.hedra.com/app/video?model=kling-ai-avatar-v2-standard) -*8 credits per second*

**Category:** Avatar\
**Audio Support:** Yes\
**Input Type:** Image + Audio

**Best For:**\
Budget avatar generation.

***

### [VEED Fabric 1.0](https://www.hedra.com/app/video?model=veed-fabric-1-0) -*20 credits per second*

**Category:** Avatar\
**Audio Support:** Yes\
**Input Type:** Image + Audio

**Best For:**\
Talking videos with expressive lip-sync.

***

# Credit Strategy Guidance

• Start with lower-cost models when testing ideas.\
• Upgrade to higher-tier models when finalizing.\
• Use Hedra-native models first for dialogue.\
• Use Motion models only when copying movement.\
• Use Premium tiers for maximum realism and narrative consistency.

| Use Case                              | Recommended Model                          |
| ------------------------------------- | ------------------------------------------ |
| On-camera dialogue, explainers, songs | Hedra Avatar                               |
| Long-form uncut video (up to 10 min)  | Hedra Avatar                               |
| Longer scenes with cuts               | Composer (combine clips from any model)    |
| Cinematic, photoreal B-roll           | Veo 3                                      |
| Fast prototyping / quick variants     | Hedra Avatar, Veo 3 Fast, or Kling 2.1 Pro |
| Action / splash / physics             | Hailou-02 Pro                              |
| Premium polished 1080p social clips   | Kling 2.1 Master                           |

<Note>
  Video AI models evolve rapidly. Each one improves often, so use this guide as directional, not permanent truth — good starting points for choosing the right fit. Always feel free to reach out to the Hedra support team if you need help.
</Note>