Today, we’re excited to share Enact, a research preview of scenes that respond to you.
Enact represents a new way of making AI video, one that’s designed to mimic the experience of directing on set. Built on Pika H3 Flex, it allows a creator to interact with “actors,” manually re-stage a scene, and manipulate the camera—all in real time.
We think this is the beginning of a new paradigm for creation. Read more and reach out for a demo below.
🥳 Ornith series models are now available on Amazon SageMaker AI.
👷♂️ With Amazon SageMaker, you can deploy Ornith 9B and 35B-A3B in a few clicks and start building for your own AI use cases.
🤗 Have fun!!
Meet Pika H3 Flex, debuting today as the #1 open weight model on DesignArena’s t2v board.
It’s built for professional iteration, with exportable audio stems, resolutions up to 4k, and faster-than-realtime generation. It can produce a 15-second clip with soundtrack in as little as 7 seconds.
So, yeah. It’s kind of a flex 💅
🚀Excited to announce Nanbeige4.3-33B!
To push the performance ceiling of small agents, our strategy in Nanbeige4.3 starts with building a powerful teacher first.
Nanbeige4.3-33B delivers SOTA agentic performance among open models at ~30B parameters:
* Strong results on agentic coding and cowork benchmarks.
* When integrated into mainstream agent harnesses, it tackles real-world tasks with cost-efficient inference.
* It will serve as a teacher for upcoming smaller Nanbeige models, starting with Nanbeige4.3-3B.
https://t.co/tSToQgZf2O
https://t.co/D228euvybz
#LLM #Agent
🐦 Ornith-1.5, including the recent 122B one, are built upon Ornith-1.0, which was developed on top of Qwen3.5 @Alibaba_Qwen with additional continued pretraining, mid-training, and post-training.
🫡 Thanks to the open-weight community!
👀 Please feel free to build your own projects or model variants on top of Ornith.
🧡Ornith-1.5-122B is now live on Hermes Agent!!!
🫡Huge thanks to @NousResearch for their extraordinary work.
Feel free to set goals for agents, give Ornith-1.5-122B a try, and let us know what you think!
🔗 https://t.co/7EEmozJAMU
We made some Unsloth UD3 dynamic quants for Ornith's new 122B release! UD-Q2_K_XL is 44.5GB and retains around 84% top-1% for wikitext general and 89% ish for chat whilst being 82% smaller (BF16 = 249GB)
Little Ornith be cooking! This time, the open weight is pretrained from scratch, achieved frontier performance, and can run on a single NVIDIA DGX. Congrats team!
🙌 Ornith-1.5-122B is live on the RunInfra API!!!
Now you can plug the model into agentic workflows via API.
🧡 Thanks @runinfrai for the support!!
🔗 https://t.co/u6BdYgTAVn
Ornith 1.5 122B is live on runinfra. launch partner with @ornith_ , live from hour 0
302 tok/s output. 72ms to first reasoning token
$0.20 / 1M input. $0.02 cached. $0.80 output
99.1% cache hit rate on agentic workloads
one of the best models we've served for its size. 10B active and it holds up on real agent work
OpenAI-compatible chat completions and Anthropic-compatible /v1/messages -- drop it into claude code as is
zero data retention by default
https://t.co/qoScb3k13i
🐦 Shoutout to @atomic_chat_hq for their expertise!
Compressed Ornith-1.5-122B from 132.6GB down to 35.0GB.
🥳 Come and try Ornith 1.5 122B on https://t.co/6zlxJ9KlcN
Congrats to @ornith_ on releasing Ornith-1.5-122B! We trained a DFlash 2 drafter for day-0 support.
It speeds up its generation by up to 3×, and it's lossless!
https://t.co/UyN3IbaS9C