{"id":6677,"date":"2026-05-02T12:56:35","date_gmt":"2026-05-02T04:56:35","guid":{"rendered":"https:\/\/crepal.ai\/blog\/?p=6677"},"modified":"2026-05-02T12:56:37","modified_gmt":"2026-05-02T04:56:37","slug":"aivideo-happyhorse-1-0-open-source","status":"publish","type":"post","link":"https:\/\/crepal.ai\/blog\/aivideo\/aivideo-happyhorse-1-0-open-source\/","title":{"rendered":"Is HappyHorse 1.0 Open Source? What’s Actually Released"},"content":{"rendered":"\n

I was halfway through a Wan 2.2 pipeline test when someone dropped a link in a Discord channel I follow: “Dora, have you seen this one?” I clicked through expecting another forgettable model card. Instead I found something sitting at #1 <\/strong>on the Artificial Analysis Video Arena \u2014 above Seedance 2.0, above Kling 3.0, above everything I’d been benchmarking for months \u2014 with no company name attached.<\/p>\n\n\n\n

My first thought wasn’t “wow.” It was: where’s the <\/em>repo<\/em>?<\/em><\/p>\n\n\n\n

So I spent the next few hours chasing download buttons and cross-referencing claims. The situation turned out to be genuinely messier than I expected \u2014 not because the model is bad, but because the gap between what’s claimed<\/em> and what’s verifiable<\/em> kept shifting. Here’s what I can actually confirm.<\/p>\n\n\n\n

The Short Answer \u2014 It’s Complicated<\/h2>\n\n\n\n

HappyHorse 1.0<\/strong> is real, Alibaba built it, and as of April 27, 2026, you can access it via API through fal.ai’s official launch<\/a>. That part is settled.<\/p>\n\n\n\n

The open-source question is not. Multiple promotional sites confidently declared full weights and commercial licensing on GitHub. fal.ai \u2014 one of the most credible sources in this story \u2014 was saying the opposite as recently as mid-April. And as of this writing, no independently verified, downloadable model weights exist under the HappyHorse name.<\/p>\n\n\n\n

Quality signal: real. API access: now live. Self-hostable open-source release: still undelivered.<\/p>\n\n\n\n

\n
\"\"<\/figure>\n<\/figure>\n\n\n\n

What the Official Sites Claim<\/h2>\n\n\n\n

Claims of full open-source release<\/h3>\n\n\n\n

Multiple sites \u2014 happyhorse-ai.com, happy-horse.art, happyhorses.io \u2014 describe the model as fully open source with commercial licensing. Claims are specific: base model weights, a distilled 8-step variant, a super-resolution module, and full inference code. One press release stated that “all model weights, distilled models, super-resolution modules, and inference code are publicly available on GitHub.”<\/p>\n\n\n\n

Base model + distilled + super-res + inference code allegedly available<\/h3>\n\n\n\n

The HuggingFace profile at happyhorse-ai\/happyhorse-1.0 lists the model as released under Apache 2.0. It describes a 15-billion parameter unified single-stream Transformer, joint audio-video generation, 7-language lip-sync, and 1080p output in roughly 38 seconds on a single H100.<\/p>\n\n\n\n

On paper: fully open, commercially usable. But the Open Source Initiative’s definition of open-source AI<\/a> requires downloadable weights, inference code, a license file, and a documented model card. Claims on a landing page don’t fulfill that checklist. Artifacts do.<\/p>\n\n\n\n

What fal.ai Has Said \u2014 And How That Changed<\/h2>\n\n\n\n

This is the part of the story with the most movement.<\/p>\n\n\n\n

The closed-source statement<\/h3>\n\n\n\n

For several weeks in April, the fal.ai HappyHorse page<\/a> carried a direct statement: “While other industry players say HappyHorse-1.0 will be open source, we can confirm that HappyHorse-1.0 will be closed source. It will not be licensable or open source.”<\/em> That language is no longer on the page as of April 27.<\/p>\n\n\n\n

fal’s blog also noted that “HappyHorse-1.0’s team has announced open-source availability with commercial licensing, but no weights or licenses have been published” \u2014 a factually accurate summary of the gap between promotional claims and actual releases.<\/p>\n\n\n\n

The official API launch<\/h3>\n\n\n\n

On April 27, fal announced it is now among the first official API partners for HappyHorse-1.0, offering four endpoints: image-to-video, reference-to-video, text-to-video, and video-edit. Commercial rights are included on all generated outputs. So the model launched as an API product \u2014 not as downloadable weights. The promotional ecosystem claiming “full open source already on GitHub” was not describing reality.<\/p>\n\n\n\n

\n
\"\"<\/figure>\n<\/figure>\n\n\n\n

What’s Actually Downloadable Right Now<\/h2>\n\n\n\n

GitHub status<\/h3>\n\n\n\n

GitHub repos under the HappyHorse name exist, but at least one is explicitly labeled a “personal information collection repository, not an official Happy Horse repository.” The official @HappyHorseATH<\/strong> account posted on April 10 that “any third-party websites or platforms claiming to offer HappyHorse services are not affiliated with the official team.” No repo with confirmed, independently reproduced weights exists under a verified Alibaba organizational account.<\/p>\n\n\n\n

HuggingFace status<\/h3>\n\n\n\n

The HuggingFace model card exists with architecture details and claims Apache 2.0 licensing. Independent verification found that as of April 10, the weights returned a 401 error. A model card without downloadable weights doesn’t let you run anything.<\/p>\n\n\n\n

Why the Confusion Exists<\/h2>\n\n\n\n

Third-party landing pages vs. Alibaba’s actual positioning<\/h3>\n\n\n\n

Alibaba’s ATH division described HappyHorse as “still under development” and in private beta when CNBC confirmed them as the creator on April 10. The official team simultaneously warned that most circulating “official websites” are fake. That vacuum was filled by third-party sites making claims the actual team hadn’t authorized \u2014 some selling credit-based access to demos, others distributing press releases announcing weights that weren’t available.<\/p>\n\n\n\n

No verified official team site yet<\/h3>\n\n\n\n

There is no single, clearly identified official Alibaba product page for HappyHorse with documentation and a verifiable GitHub organization. That’s the structural problem, and it’s what made the open-source question so hard to answer cleanly.<\/p>\n\n\n\n

\n
\"\"<\/figure>\n<\/figure>\n\n\n\n

What This Means for You<\/h2>\n\n\n\n

If you want to self-host<\/h3>\n\n\n\n

Don’t build anything around HappyHorse today on the assumption that weights are available. The leaderboard<\/a> numbers are real \u2014 T2V Elo 1,357, I2V Elo 1,403, with a 74-point lead over Seedance 2.0 that’s the largest margin in Artificial Analysis history. But no independently confirmed, downloadable weight file exists for local deployment.<\/p>\n\n\n\n

For a model you can actually run today, Alibaba’s own Wan 2.6 release<\/a> is the honest comparison. Published in December 2025 under Apache 2.0, with real weights on HuggingFace, functional inference code, active community deployment, ComfyUI wrappers, and quantized variants. Its T2V Elo sits around 1,189 \u2014 lower than HappyHorse \u2014 but you can download it, fine-tune it, and build production pipelines today. That gap between a model you can evaluate and one you can only read about doesn’t close until weights actually ship.<\/p>\n\n\n\n

If you just want to generate videos<\/h3>\n\n\n\n

The API is live on fal.ai as of April 27 \u2014 four endpoints, 720p and 1080p output, commercial rights included. Verify current pricing before committing to a production workflow, since early-launch terms shift.<\/p>\n\n\n\n

How This Compares to Other Chinese Open-Source Video Models<\/h2>\n\n\n\n

Wan 2.6 \u2014 what a real open-source release looks like<\/h3>\n\n\n\n

Wan 2.6 is the reference point for what “open source” actually means in practice. Downloadable weights under Apache 2.0, third-party integrations, ComfyUI nodes, community forks, quantized versions \u2014 the full stack exists and works. It’s a 14-billion parameter Mixture-of-Experts model with multi-shot generation, native audio sync, and up to 15-second clips.<\/p>\n\n\n\n

HappyHorse has a stronger arena score. Wan 2.6 has an actual release. If the question is “which can I build on today,” the answer is still Wan 2.6 \u2014 by a wide margin.<\/p>\n\n\n\n

Longcat Video \u2014 the architecture question<\/h3>\n\n\n\n

Some community analysis has linked HappyHorse’s underlying architecture to the daVinci-MagiHuman project from Sand.ai and GAIR Lab, which is genuinely open source with Apache 2.0 weights on HuggingFace. A 36Kr investigation found near-identical benchmark performance and similar site structures between the two. That connection remains unconfirmed by either team \u2014 but it’s the most credible theory about HappyHorse’s technical lineage, and worth knowing if you’re evaluating what “open source” might eventually look like if weights do ship.<\/p>\n\n\n\n

\n
\"\"<\/figure>\n<\/figure>\n\n\n\n

What to Watch For Next<\/h2>\n\n\n\n

Three signals matter:<\/p>\n\n\n\n

Weights published to a verified Alibaba <\/strong>GitHub<\/strong> organization.<\/strong> Not a third-party repo with unconfirmed provenance. A real, signed release from the team \u2014 that’s the trigger for self-hosting to become possible.<\/p>\n\n\n\n

Independent reproduction.<\/strong> Someone outside the creators downloads weights, runs inference, publishes results. Until that happens, architecture specs remain self-reported, not verified.<\/p>\n\n\n\n

API<\/strong> stability and pricing transparency.<\/strong> The fal.ai API is live as of April 27. For most creators, that’s the practical access path right now. Verify current pricing and rate limits directly \u2014 early-launch terms frequently shift.<\/p>\n\n\n\n

Conclusion<\/h2>\n\n\n\n

HappyHorse 1.0 is real. The leaderboard performance is real \u2014 blind human votes don’t lie about which video looks better. Alibaba built it, fal.ai launched API access on April 27, and the arena lead over Seedance 2.0 is historically large.<\/p>\n\n\n\n

But “open source” has a specific meaning, and HappyHorse hasn’t met it yet. The API is open. The weights are not. Promotional sites calling it “fully open source with weights on GitHub” were describing an intention, not a verified release.<\/p>\n\n\n\n

I’m watching for an official Alibaba GitHub release with a real license file. That’s when the story actually changes. Until then \u2014 great model, messy release, use the API if you need it now.<\/p>\n\n\n\n

\n\n\n\n

Previous Posts:<\/strong><\/p>\n\n\n\n

\n
What Is HappyHorse-1.0? What AI Video Creators Should Know<\/a><\/blockquote>