Pika releases open-weight H3 Flex video model with audio stems
READ THE ORIGINAL SOURCEOpen original sourcepika.art · Opens in a new tabPika announced H3 Flex, a post-trained version of MiniMax H3 for video generation. The model is distilled to six steps instead of 50, accepts up to 12 reference files, generates synchronized stereo audio, and offers five separate audio stems via API. Drafts generate at 480p or 768p, with an upsampler reaching 1080p, 2K, or 4K. Generator adapters and inference code are released as open weights, while the upsampler, stem separation, and serving stack remain Pika-hosted. The company reports a DesignArena text-to-video ranking at launch.
One primary source; other links are context, not independent corroboration.
- Primary sourceX discovery · automated brief
- Discovery linkX
Grok Bot submitted the link. AINewsIQ fetched the primary source and prepared a source-based brief automatically. An editor has not reviewed the story; social attention and the brief do not independently verify its claims.
Open full article view →- Linked article+30
- Fetched source page+30
- No editor link review+0
- Original non-discussion link+5
This measures the stored source trail and editorial steps. It does not certify accuracy, peer review, or independent corroboration. Popularity adds no points.
View full scoring →Read the scoring rubric →What the source says
Pika announced H3 Flex, a post-trained version of MiniMax H3 for video generation. The model is distilled to six steps instead of 50, accepts up to 12 reference files, generates synchronized stereo audio, and offers five separate audio stems via API. Drafts generate at 480p or 768p, with an upsampler reaching 1080p, 2K, or 4K. Generator adapters and inference code are released as open weights, while the upsampler, stem separation, and serving stack remain Pika-hosted. The company reports a DesignArena text-to-video ranking at launch.
A READER'S EVIDENCE CHECKHow to read this signalSix questions about claims, sources, risks, and open checks
This is a signal-level guide until claim-level evidence is available. “Not recorded” means the evidence has not been assessed here; it does not mean a result is false. How we check AI news →
- 01Who is making each claim?
The claimant and whether each claim was measured or asserted are not recorded in this public brief.
- 02What was actually tested?
Not recorded per claim. Check the source for sample size, tasks, model and benchmark versions, and reproducibility.
- 03Where are the sources?
Original link: pika.art. Brief basis: fetched source page. Primary page date and quote locations are not established in this panel.
- 04What could mislead?
Source claims not independently checked; the seven-second speed figure is a controlled transformer-computation benchmark, not end-to-end latency, and the leaderboard rank is a launch snapshot.
- 05What is at risk if wrong?
Not assessed per claim. Consider safety, privacy, cost, and adoption decisions before relying on the result.
- 06What would change the view?
No claim-specific check is recorded. An inspectable primary method or independent replication could change the assessment.
Key findings from the stored brief
- Flex post-trains MiniMax H3 to generate in six steps instead of 50, then refines it for quality and control.
- The model accepts up to 12 reference files combining images, video, and audio in one generation.
- Open weights release covers text-to-video, first/last-frame, and reference-to-video at 480p and 768p; upsampling and stem separation are Pika-hosted.
Source claims not independently checked; the seven-second speed figure is a controlled transformer-computation benchmark, not end-to-end latency, and the leaderboard rank is a launch snapshot.
This is a source-based abstract, not independent validation of the source's claims.
Evidence & sources
These links show where the signal came from and what the source page linked to. A linked page is context, not independent corroboration.
Related archive coverage
These relationships record a citation or topic overlap, not confirmation of the finding.
Grok Bot submitted the link. AINewsIQ fetched the primary source and prepared a source-based brief automatically. An editor has not reviewed the story; social attention and the brief do not independently verify its claims.
These are Grok Bot discovery notes. AINewsIQ prepared a brief from available source material. The bot's rationale, linked posts, authors, and stances have not been independently verified.
Pika released H3 Flex, its post-trained MiniMax H3 video model, with open weights. Pika says it can generate a 15-second 768p clip with synced audio in about seven seconds of transformer compute, takes up to 12 references, exports five audio stems and finishes at up to 4K. DesignArena ranked it #7 on its video board.
Source claims not independently checked; the seven-second speed figure is a controlled transformer-computation benchmark, not end-to-end latency, and the leaderboard rank is a launch snapshot.
- Bot tagged: supports DesignArena says Pika H3 Flex reached #7 on its Video leaderboard with a 1293 Elo rating. Inspect post ↗
Engagement counts, when shown, are observed snapshots. They do not establish accuracy or corroboration.
- x post ↗ · 1,010 likes, 205 reposts, 191 replies, 176 quotes · observed 10/30/2026, 4:10:51 AM
Timeline & updates
- Signal dated by source or discovery feed
This timestamp places the signal in the radar. It may be the discovery post date rather than the primary page's first publication date.
- Shared brief stored
AINewsIQ prepared one abstract from fetched source page.
A publication revision history is not stored for this signal. New engagement snapshots do not mean the source itself changed.
Why this may matter
Open-weight video models with reference conditioning and audio generation give developers and researchers a locally runnable option for creative video tools, though key high-resolution and audio-separation features remain behind Pika's hosted service.
Limits and open questions
Source claims not independently checked; the seven-second speed figure is a controlled transformer-computation benchmark, not end-to-end latency, and the leaderboard rank is a launch snapshot.
Organization-specific financial, security, and operational effects require the organization's own context and review.
Discussion
Comments are reviewed before publication. Discuss the evidence, not the person.
No published comments yet.
Reader discussion opens with accounts. You can read the original source and published comments now.