Two athletes playing table tennis beside a sunlit pool

PikaH3FLEX

Our faster-than-realtime model.
Built for professional iteration.

Use API

Pika H3 Flex

Iteration, without the wait.

Pika H3 Flex is a post-trained H3 model designed to give professional creators the speed to explore and the control to be precise. It keeps the quality you need while generating faster than real-time in almost every configuration.

Resolutions reach 4K, references reach twelve inputs, and editing happens through a reference plus a prompt. Exportable audio stems let you isolate and refine any layer of the soundscape.

Speed and price

Speed and price

Pika H3 Flex1.8×

faster than H3Max

Pika H3 Flex80%

cheaper at matched quality

On average, across matched text-to-video, image-to-video and reference-to-video comparisons. DiT time, not guaranteed end-to-end latency.

Iterate
faster than
real-time

A 15 second clip in as little as 7 seconds. Creation at the speed of thought, so the next idea is always one take away.

*Best-case performance.

Built for tinkering

Like a pro.

Fast drafts are only useful when they can become final shots. Flex keeps the take, the audio and every reference intact from the first pass to the finish.

A low-resolution draft of the poolside table tennis scene768p draft
The same scene finished in 4K4K finish
01

Draft at 768p, finish at 4K

Explore at lower resolution, then upscale to 1080p, 2K or 4K in finishing mode. The take, framing and audio are preserved.

  • 768p
  • 1080p
  • 2K
  • 4K
02

Build from multiple references

Combine sketch, image, sound, video and moodboards in one request. Up to twelve inputs hold the look together.

03

Add, remove or swap anything

Start with a clip and edit with a reference or a prompt. Change a character, camera move, voice or look without rebuilding the shot.

Separate audio stems

Every layer,
on its own track.

Export up to three audio stems to isolate and refine any facet of a clip's soundscape: dialogue, music and effects.

5s10s15s
Dialogue
Music
Effects
3 stems · stereo 32 kHzWAV

A flexible performer

One model,
many kinds of motion.

Flex is built for a wide range of video types and genres, from product films to animated shorts.

01

Complex camera moves

02

Multi-shot

03

Motion graphics

04

Product films

05

Animated shorts

Direct anything from orbits to dollies, cranes, tracking shots, whip pans, macro moves and changing lenses in natural language.

Make it move.

Quality

Ranked #7
in video quality.

Flex sits above base H3, H3Max and Seedance 2.5 on DesignArena's text-to-video board, and it is the highest ranked model with open weights.

DesignArena leaderboard snapshot · open weights highlighted
  1. 01Gemini Omni 1.1 Flash1360
  2. 02Gemini Omni Flash1343
  3. 03HeyGen Video 1.01334
  4. 04Wan 3.01324
  5. 05P-Video-2-Pro1310
  6. 06FLUX 3 Max1297
  7. 07Pika H3 Flex1293
  8. 08FLUX 3 Video1286
  9. 09MiniMax H3 Max1277
  10. 10MiniMax H31277
Launch promo

15 bonus generations per day

For active Pika H3 Flex members. Until November 15th.

See plans

Priced to play

Pay for seconds,
not subscriptions you never use.

Text, image and keyframe generation are charged per output-video second. Draft-to-video finishing charges only the difference between the draft and the finish.

OutputAPI regularAPI launchMember price
480p$0.02$0.011 credit
768p$0.04$0.022 credits
1080p$0.07$0.0353.5 credits
2K$0.08$0.044 credits
4K$0.16$0.088 credits
1 Flex credit = $0.01Compare memberships

Open weights

Built in the open.

Pika H3 Flex is released with open weights and an inference implementation. Local inference will not necessarily match the optimised hosted API, but the model is yours to inspect, run and build on.

Open weights · research-friendly release
01

Model card

Weights, licence and evaluation notes

Included
02

Inference kit

Reference implementation and setup

Included
03

ComfyUI nodes

A visual workflow for the model

Included

Full model specs

The details.

Base model
Post-trained on MiniMax H3
Generation modes
Text to video, first/last-frame image to video, multimodal reference to video, prompt + reference-guided editing
Resolution
480p, 768p, 1080p, 2K, 4K
Duration
3–15 seconds in whole-second requests
Frame rate
24 fps
Aspect ratios
21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16, 9:21
Native audio
Stereo, 32 kHz
Audio stems
Dialogue, Music and Effects WAV files returned beside the finished video
Reference limits
12 assets: up to 9 images, 3 videos and 3 audio clips
Prompt length
Up to 8,000 characters
Seed control
Supported
Output format
MP4 with H.264 video and AAC audio
Open weights
Hugging Face weights plus a GitHub inference implementation

Get started

Try Pika H3 Flex

Or just enjoy the video. Then try it.

Use API

FAQ

Good questions.

Everything else lives in the specs above. For billing, see the pricing page or your account settings.

01What is Pika H3 Flex?

Pika H3 Flex is a post-trained H3 video model built for professional iteration. Text, image, reference and edit workflows share one model, so a draft can become a finished shot without changing tools.

02How fast is Pika H3 Flex?

In its fastest configuration it can generate a 15 second clip in as little as 7 seconds. Across matched text, image and reference comparisons it averages 1.8x faster than H3Max.

03Does “seven seconds” mean my request always finishes in seven seconds?

No. Seven seconds is the best-case generation time for a 15 second clip. Queue depth, resolution, references and finishing mode all affect end-to-end latency.

04What can I use as a reference?

Up to 12 assets in total: images, video clips, audio, sketches, moodboards or a written prompt. At least one image or video is required in reference mode.

05Can I change part of a clip without starting over?

Yes. Start from an existing video and add a reference or prompt to adjust a character, camera move, look or voice while keeping the rest of the take.

06Does Pika H3 Flex generate in 4K?

Yes. The request generates at 768p and upscales once inside the same job to 1080p, 2K or 4K, preserving the generated image content and audio.

07How do separate audio stems work?

When requested, every generation endpoint can return the video plus up to three WAV files for dialogue, music and effects.

08What is included in the open weights release?

The model weights and an inference implementation are published together. Local inference speed depends on your hardware and optimisation choices.

09Where can I try Pika H3 Flex?

Use any Try Pika button on this page. You will be asked to sign in with Google, choose a paid plan and then enter the Studio queue.