Drop files here, paste files, or add a URL. Photo mode: one photo of the person to animate. JPEG, PNG, or WebP.
Drop files here, paste files, or add a URL. Optional. Adding a video switches to reference video mode: an animated or illustrated character, at least 2 seconds long and up to 50 MB.
Drop files here, paste files, or add a URL. Speech audio, 2–60 seconds and up to 15 MB. The video runs the length of the audio. MP3, WAV, M4A, AAC, or OGG.
Example Output
This is sample data. Generate a new result to see real output.
Pay as you go. No subscription, no hidden fees.
| Tier | 1080p | 2K |
|---|---|---|
| Avatar 6 | $0.080/sec 16 credits/sec $0.100/secSave 20% | $0.116/sec 23.2 credits/sec $0.145/secSave 20% |
| Avatar 6 Turbo | $0.040/sec 8 credits/sec $0.050/secSave 20% | $0.070/sec 14 credits/sec $0.087/secSave 20% |
Charges are based on the final output.
Photo, character video, or stock avatar to lip-synced talking video in 1080p or 2K
HeyGen Avatar 6 API: Turn a Photo into a Talking Video
HeyGen Avatar 6 API gives you HeyGen's newest talking-avatar model, released on November 17, 2026: it turns one photo, a cartoon character video, or a HeyGen stock avatar into a lip-synced talking video in 1080p or 2K, with gestures that follow the speech. Choose heygen-avatar-6 or heygen-avatar-6-turbo at half the 1080p price.
Available HeyGen Avatar 6 API Models on PoYo
HeyGen Avatar 6 Turbo (heygen-avatar-6-turbo)
Key Features of HeyGen Avatar 6 API
The HeyGen Avatar 6 API offers three ways to make a talking avatar with one request format: send a photo with audio, a character video with audio, or a stock avatar with text or audio, then direct the gestures with a prompt.
01
Turn a Photo into a Talking Video in 1080p or 2K
The HeyGen Avatar 6 API turns one photo of a person into a talking photo video: add your speech audio and it returns a lip-synced clip that runs as long as the audio, from 2 to 60 seconds. Send the photo in image_urls and the speech in audio_url; the video keeps the photo's aspect ratio. In PoYo tests, a square photo with 5 seconds of audio returned a 1088×1088 video at 1080p in about 48 seconds. In this HeyGen sample, a single photo becomes an office presenter who talks and gestures naturally.
- One photo in image_urls plus 2–60 s of audio
- Video length follows the audio track
- 1080p by default, 2K when you set resolution
02
Animate Cartoon and Illustrated Characters from a Reference Video
With the HeyGen Avatar 6 API, mascots, animated hosts, and illustrated characters can speak too. Send a video of the character in video_url, at least 2 seconds long and up to 50 MB, with speech in audio_url, and add a reference photo of the same character in image_urls if you have one. In this PoYo example, a 5-second greenhouse clip, the character's still image, and an 8.3-second voice line produced an 8.2-second 1080p video, with a motion prompt that cues a wave on "Hi".
- Reference video of a cartoon or illustrated character
- Optional photo of the same character in image_urls
- Video keeps the reference clip's aspect ratio
03
1,248 HeyGen Stock Avatars That Read Your Text
The HeyGen Avatar 6 API can start a presenter video from a script alone. Pick one of 1,248 HeyGen stock avatar looks across 156 presenters in avatar, then send up to 1,000 characters in text with one of 102 listed voices, or send your own audio_url instead. Choose 16:9, 9:16, 1:1, 4:5, or 5:4. In this PoYo example, Annie Sofa Sitting Front read a 29-word script in the Ivy voice and returned a 10.8-second 1080p video, billed as 11 seconds.
- Avatar names match regardless of letter case
- Text requests are billed by the actual video length
- Five aspect ratios in public avatar mode
04
Motion Prompts That Cue Gestures on Specific Words
In the HeyGen Avatar 6 API, write the action next to the word, and the avatar performs it on cue. The optional prompt field takes up to 1,200 characters and works in all three modes. HeyGen's sample used: "Friendly pharmacist. On 'Hello', give a thumbs up. On 'the counter on the left', point to the side with your index finger. On 'always happy to chat', place your right hand on your heart. On 'some water', wave your right hand." The presenter gives each gesture at the matching word.
- Up to 1,200 characters of direction
- Tie gestures and expressions to exact words
- Same field for photo, video, and avatar modes
05
Gestures That Follow What Is Said
Even without a prompt, HeyGen Avatar 6 follows the meaning of the speech: according to HeyGen, the model changes gestures and expressions line by line, such as pointing, counting, or directional motion, rather than repeating one motion. It works the same with audio or text input. In this HeyGen sample, the presenter's expression shifts with each sentence.
- Gestures matched to the script's meaning
- Expressions change from line to line
- Works with audio or text input
Who Can Benefit from HeyGen Avatar 6 API?
E-Learning and Training Teams
Turn lesson scripts into presenter videos with the HeyGen Avatar 6 API, a stock avatar, and a listed voice, then regenerate a module when the text changes. Each clip runs to the length of the narration, and Turbo keeps drafts at 8 credits per second at 1080p.
Marketing and Advertising Teams
Produce product explainers and ad variants in 16:9, 9:16, or 1:1 from one script. Test hooks with HeyGen Avatar 6 Turbo, then render the final cut with HeyGen Avatar 6 in 2K.
Sales and Customer Outreach
Send prospects short personalized videos: animate a rep's photo with a recorded message, or let a stock presenter read a tailored script. A 20-second 1080p Turbo clip costs 160 credits ($0.80).
Localization Teams
Pair the same photo or stock avatar with a voice-over for each market, so every version keeps one presenter. Each clip follows the audio you supply, from 2 to 60 seconds.
Creators and Animation Studios
Give mascots and illustrated characters a voice from a short reference clip, and use the motion prompt to place a wave or a nod on the right word, without animating mouth shapes by hand.
AI Video Apps and SaaS Platforms
Add a talking avatar API to your product: photo to talking video and text-to-presenter features through the HeyGen Avatar 6 API: one asynchronous request format with two model IDs, per-second pricing, and callback_url notifications.
HeyGen Avatar 6 vs Avatar 6 Turbo — Which One Fits Your Project?
The two models of the HeyGen Avatar 6 API share the same request fields, modes, and limits on PoYo, so switching only means changing the model ID. The difference is price and HeyGen's speed and quality tier: HeyGen Avatar 6 Turbo costs half at 1080p and runs on HeyGen's lower-latency stack.
| Feature | HeyGen Avatar 6 | HeyGen Avatar 6 Turbo |
|---|---|---|
| Model ID | heygen-avatar-6 | heygen-avatar-6-turbo |
| Best for | Final presenter and character videos | Drafts, previews, and high-volume clips |
| HeyGen tier | Avatar 6 Quality | Avatar 6 Turbo, on HeyGen's accelerated inference stack |
| Price at 1080p | 16 credits ($0.08) per second | 8 credits ($0.04) per second |
| Price at 2K | 23.2 credits ($0.116) per second | 14 credits ($0.07) per second |
| 10-second 1080p video | 160 credits ($0.80) | 80 credits ($0.40) |
| Input modes | Photo + audio; character video + audio; stock avatar + text or audio | Same three modes |
| Output resolution | 1080p (default) or 2K | 1080p (default) or 2K |
| Driving audio | 2–60 seconds, up to 15 MB | 2–60 seconds, up to 15 MB |
| Stock avatars and voices | 1,248 looks; text up to 1,000 characters; 102 listed voices | Same list and limits |
| Motion prompt | Up to 1,200 characters | Up to 1,200 characters |
| Aspect ratio in avatar mode | 16:9, 9:16, 1:1, 4:5, 5:4 | 16:9, 9:16, 1:1, 4:5, 5:4 |
| PoYo test runs | Photo, 1080p, 4.9 s video: about 48 s. Stock avatar + text, 2K, 5.2 s video: about 3 min | Stock avatar + text, 1080p, 14.6–18.6 s videos: about 1.7–1.9 min. Character video, 2K, 4.9 s video: about 2.7 min |
| Billing | Audio: rounded-up audio seconds, charged at submission. Text: rounded-up seconds of the finished video | Same rules |
Prices and limits are for the HeyGen Avatar 6 API models heygen-avatar-6 and heygen-avatar-6-turbo on PoYo as of November 18, 2026. Test runs are single PoYo test-environment tasks and vary with clip length and load.
How to Use HeyGen Avatar 6 API on PoYo
Step 1: Create Your PoYo API Key
Sign in to PoYo and create an API key in the dashboard; the same key works for both HeyGen Avatar 6 API models, heygen-avatar-6 and heygen-avatar-6-turbo.
Create API Key →
Step 2: Add Credits
Top up your balance. A 10-second 1080p video costs 160 credits with HeyGen Avatar 6 or 80 credits with HeyGen Avatar 6 Turbo.
Add Credits →
Step 3: Submit a Task and Get the Video
To call the HeyGen Avatar 6 API, send a POST request to /api/generate/submit with "model": "heygen-avatar-6" and one mode's inputs: image_urls + audio_url for a photo, video_url + audio_url for a character video, or avatar + text (or audio_url) for a stock avatar. Poll the returned task_id or set a callback_url. The HeyGen Avatar 6 API documentation lists every avatar and voice name.
Read the API Docs →
Frequently Asked Questions about HeyGen Avatar 6 API
What is HeyGen Avatar 6?
HeyGen Avatar 6 is HeyGen's talking-avatar video model, released on November 17, 2026. It animates a photo, a cartoon or illustrated character video, or a HeyGen stock avatar into a lip-synced talking video in 1080p or 2K, with gestures matched to the speech. On PoYo, the HeyGen Avatar 6 API offers heygen-avatar-6 and the lower-latency heygen-avatar-6-turbo, with pay-per-second pricing and a Playground for testing.
Is there a HeyGen API for Avatar 6?
Yes. HeyGen offers Avatar 6 through its own HeyGen API, and on PoYo you call the HeyGen Avatar 6 API with a POST request to /api/generate/submit using the model ID heygen-avatar-6 or heygen-avatar-6-turbo, then read the result from the task status endpoint or a callback_url. One PoYo API key covers both model IDs and every other model on PoYo, and the API documentation includes a request example for each input mode.
How much does the HeyGen Avatar 6 API cost?
For HeyGen API pricing on PoYo, HeyGen Avatar 6 costs 16 credits ($0.08) per second at 1080p and 23.2 credits ($0.116) at 2K. HeyGen Avatar 6 Turbo costs 8 credits ($0.04) per second at 1080p and 14 credits ($0.07) at 2K. With audio, the charge is the audio length rounded up to whole seconds; with text, it is the finished video's length rounded up. A 10-second 1080p Turbo video costs 80 credits ($0.40), and you pay only for the seconds you generate, with no subscription.
How do I get a HeyGen Avatar 6 API key?
Create a PoYo account, open the API Keys page in the dashboard, and create a key. Add credits on the billing page, then send the key as a Bearer token with your request. The same PoYo key works for both HeyGen Avatar 6 API models, heygen-avatar-6 and heygen-avatar-6-turbo, and you can try both in the Playground on this page before writing code.
What is the difference between HeyGen Avatar 6 and Avatar 6 Turbo?
Both take the same inputs, limits, avatars, and voices and output 1080p or 2K. HeyGen Avatar 6 is HeyGen's Quality tier; HeyGen Avatar 6 Turbo runs on HeyGen's accelerated inference stack for lower latency and costs half at 1080p (8 versus 16 credits per second). Start with Turbo for drafts and switch the model ID for final renders.
How do I turn a photo into a talking video with the HeyGen Avatar 6 API?
Send one photo of the person in image_urls and speech audio of 2–60 seconds, up to 15 MB, in audio_url. The HeyGen Avatar 6 API returns a lip-synced video as long as the audio, in the photo's aspect ratio. Add an optional prompt to direct gestures. In a PoYo test, a square photo with 5 seconds of audio returned a 1088×1088 1080p video in about 48 seconds, billed as 6 seconds.
Can the HeyGen Avatar 6 API animate cartoon or illustrated characters?
Yes. Reference video mode is made for animated and illustrated characters: send a video of the character, at least 2 seconds long and up to 50 MB, in video_url, plus speech in audio_url, and optionally a photo of the same character in image_urls. For real people, use photo mode with one photo in image_urls. The output keeps the reference video's aspect ratio.
Which stock avatars and voices can I use with the HeyGen Avatar 6 API?
The HeyGen Avatar 6 API accepts 1,248 HeyGen stock avatar looks across 156 presenters, such as Abigail Sofa Front or Annie Sofa Sitting Front, and names match regardless of letter case or extra spaces. For text, choose one of 102 listed voices, such as Warm Pro Narrator (the default), Ivy, or Chill Brian. The full lists are in the API documentation.
Can the avatar read text instead of audio?
Yes, in public avatar mode. Send avatar with up to 1,000 characters in text and an optional voice, and HeyGen Avatar 6 speaks the text in that voice. The charge follows the finished video's length, rounded up to whole seconds, and submitting needs credits for about 1.5 times the estimated reading time. A 29-word English script produced a 10.8-second video in a PoYo test.
How long can a HeyGen Avatar 6 API video be?
With audio, each video runs as long as the driving audio, from 2 to 60 seconds, and the finished clip is about 0.1–0.2 seconds shorter than the audio. With text in public avatar mode, the length follows the reading time of up to 1,000 characters; in PoYo tests, about 300 characters of English produced 18.2–18.6 seconds of video.
What resolutions and aspect ratios does the HeyGen Avatar 6 API support?
Set resolution to 1080p (the default) or 2K. In photo and reference video modes, the video keeps the aspect ratio of your input; in PoYo tests, a square photo returned 1088×1088 at 1080p and 1632×1632 at 2K. In public avatar mode, choose 16:9 (default), 9:16, 1:1, 4:5, or 5:4; a 16:9 2K video measured 2560×1440.
Can I control gestures and expressions in the HeyGen Avatar 6 API?
Yes. The optional prompt field takes up to 1,200 characters of direction for gestures, expressions, and posture, and you can tie an action to a word, for example "On 'Hello', give a thumbs up." Without a prompt, HeyGen Avatar 6 picks gestures that match the meaning of the speech, such as pointing or counting, according to HeyGen.
How long does a HeyGen Avatar 6 API generation take?
In PoYo test-environment runs, photo and reference video tasks finished in about 45 seconds to 3 minutes, and public avatar tasks in about 1 to 4 minutes, depending on length, resolution, and load. A 4.9-second 1080p video from a photo took about 48 seconds. Use callback_url or poll the task status instead of waiting on the request.
How is HeyGen Avatar 6 different from Avatar V?
According to HeyGen, Avatar V reused the same gestures, while Avatar 6 reads the script and gestures to match it, with more realistic lip sync, native 2K frames, and motion prompts that place an action on a specific word. HeyGen Avatar 6 also comes in a lower-latency Turbo tier. On PoYo, the HeyGen Avatar 6 API offers both tiers through one request format.
How does HeyGen Avatar 6 compare with OmniHuman 2.0 on PoYo?
Both animate a portrait with audio on PoYo. The HeyGen Avatar 6 API outputs 1080p or 2K from 2–60 seconds of audio, adds HeyGen stock avatars that read text and a reference video mode for animated characters, and accepts word-level motion prompts. OmniHuman 2.0 outputs 720p from 4–300 seconds of audio and suits longer speeches or songs. Pick by length, resolution, and input type.
Can I use the HeyGen Avatar 6 API with Python or other languages?
Yes. The HeyGen Avatar 6 API on PoYo is a standard HTTPS JSON API, so any HTTP client works, including Python requests, Node.js fetch, and cURL. Send the request to /api/generate/submit with your API key, then poll /api/generate/status/{task_id}. The API documentation shows the request body for photo, reference video, avatar with audio, and avatar with text.
Why Use PoYo for HeyGen Avatar 6 API Access
Two Tiers, One Request Format
Switch between the two HeyGen Avatar 6 API models by changing the model ID; fields, modes, avatars, and voices stay the same, and the same key works across PoYo's video models.
Per-Second Pricing with Published Rules
Pay by output second at 8 to 23.2 credits, with audio requests billed by rounded-up audio length and text requests by the finished video, and see the estimate before you run.
Playground with Real Examples
Try photo, character video, and stock avatar modes in the Playground, pick avatars and voices from searchable lists, and copy the exact JSON request into your code.
Simple Async Integration
Submit a task, then poll the status endpoint or receive a callback_url notification; finished videos are returned as downloadable MP4 links.







