This week, the next generation of OctaneRender was released, featuring an all new neural rendering platform that synthesizes 3D and AI creation workflows into one unified pipeline — with a bridge to unlimited GPU compute power on the Render Network.
In tandem, the Render Network added two more leading text-to-image generators to the AI generation portal: Flux by Blackforest Labs, and Dream Machine by Luma Labs. Both tools are live on the Render Network today, available for use with RENDER Credits, joining Stable Diffusion as new AI services on the network. Flux is now also the first text-to-video generative AI model implemented on the Render Network, with others including RunwayML coming soon.
This blog will break down the recent updates as well as recap major milestones, talks, and initiatives for the Render Network this fall.
The Next Generation of OctaneRender is Introduced Featuring Deepened Render Network Integration
This week OTOY introduced future versions of OctaneRender, in its annual Black Friday sale, featuring major updates for the Render Network.
Future versions of Octane showcased a new AI module, bringing leading generative AI text to image tools to the Octane Node Graph, including: Topaz AI, RunwayML, LumaLabs, FLUX, Skybox 360, Kling, MinMax and other popular services — with streamlined distribution to Render.
New Alpha builds of Octane also feature a new Render Network Browser, that enables artists to dock the Render Network portal like any other work pane or keep it in a separate window to check uploaded scenes and render jobs directly from the Octane interface. With AI tools supported inside in Octane AI modules and a Render Network browser, Alpha versions of Octane feature tighter coupling of 3D and AI creative workflows, and a more integrated connection to Render’s compute platform.
Finally, the Octane Alpha update previewed a python-based launcher designed to streamline distribution of generative AI jobs to the network’s near unlimited GPU compute power. The update included plans to run generative AI services like Topaz AI, Skybox 360, and leading text-to-image generators on the Render Network’s decentralized GPU nodes, providing cost and scale efficiencies to augment 3D and AI workflows.
In addition to integrating 3D tools into the Render Network, Octane also featured a new neural rendering architecture with all new features built for the future of AI accelerated holographic rendering and spatial computing. These include features presented in Alpha builds of Octane like: Meshlets, Virtual Texture Streaming, Neural Radiance Caching, and 3D Gaussian Splatting, making it easier to create fully immersive holographic ‘neural objects’ using Octane, with augmentation from the Render Network.
Finally, as part of the OctaneStudio+ Black Friday release, all annual Octane subscribers are getting up to €200 in Render Credits to try integrating the Render Network into their pipelines this year, and in 2025 — encouraging all Octane subscribers to take advantage of unlimited compute on Render.
These updates to Octane, make the tool a hub for creative workflows spanning 3D and AI, augmented by frictionless access to near unlimited high performance compute on the Render Network.
Advances to the Render Network
This week’s Octane update complements accelerating progress in the development of the Render Network, with a number of major new features and product advancements released over the past few months, including:
Native Cinema 4D Support (Out of Beta)
After extensive testing Native Cinema 4D file support moved out of Beta earlier this year, enabling artists from the most widely used motion graphics pipeline to submit jobs to the Render Network without having to go through cumbersome importing and exporting steps, making the network much easier to use in production workflows.
Render Network Cinema 4D Wizard
Proposed and passed in RNP-012, the Render Network introduced a new Cinema 4D Wizard providing an application to simplify sending jobs to the Render Network, check scenes for any errors prior to rendering them, and providing a push button artist friendly interface for getting jobs on Render.
Render Network Redshift Support (Beta)
This fall, the Render Network expanded to support Redshift, making both of the industry’s leading GPU render engines now supported on the Render Network. With this integration, users from the industry’s leading GPU rendering pipelines can access near unlimited compute on Render, with the ability to easily mix and match render engines — across biased and unbiased rendering — to meet a project’s desired look.
Render Network Support for Cycles Enters Beta
Continuing on the network’s multi-render expansion to support more render engines — starting with Redshift — in October, Blender’s Cycles entered a closed beta on the Render Network, beginning support for the GPU render engine of choice in the largest 3D ecosystem, Blender. Already artists using the Beta are showing great results, promising to bring high performance GPU cloud compute to the most widely used 3D pipeline.
Dropbox and AWS S3 Cloud Storage Integrations
With the goal of making studio production workflows better supported on the Render Network, new cloud storage integrations with Dropbox and AWS S3 were released, enabling artists and studios to send jobs from the Render Network to cloud storage providers without having to download and reupload frames. This helps with large rendering workflows and collaborative production workflows, where many parties need easy access in real time to downloaded frames.
Finally, a Download Manager Application enables artists to customize large-scale downloading preferences, making studio production workflows better supported on the network. All of these features make the Render Network much more user friendly for the leading 3D ecosystems Cinema 4D and Blender, as well as better support more intensive production use of the network by studios and production houses.
Render Network Talks and Keynotes
Throughout the fall, the Render Network took part in a number of prominent talks and panels, to share community updates and discuss the network’s progress in shaping the future of decentralized computing and immersive media.
TechForum Argentina
In a keynote talk to industry luminaries and heads of state at TechForum Argentina, Founder of the Render Network, Jules Urbach, presented a showcase of how the Render Network is building decentralized computing infrastructure for the era of holographic media and spatial computing. The talk showcased glimpses into expansions to The Archive, which were on display earlier this month with the release of “765874 — Unification”.
Blockworks Permissionlesss — Crytpo x AI
Founder of the Render Network Jules Urbach joined a Main Stage panel at Blockworks’s Permissionless on Crypto x AI with Casey Caruso, Founder of Topology and SamHogan from kuzco.xyz. During the talk, Jules focused on how the Render Network is focused on a transformation within the field of generative AI towards “inference time compute”.
Jules discussed how compute needs for large language models (LLMs) are transitioning from foundation model training towards inference and fine tuning processes, that are well served with consumer GPUs. He described how open source LLMs are becoming increasingly powerful, making pre-training new models impractical for most generative AI applications.
Rather than spending compute resources on new model development, Jules pointed to studies showing that ‘large context window’ inference intensive operations are leading to better outputs and performance gains. Jules also noted how the rise of LoRAs enable artists and developers to fine tune open source models with their own unique data sets, making building on top of base models more effective than training new foundation models.
Throughout the talk Jules discussed how DePIN networks like Render are well served to democratize the field of AI by combining the performance gains of inference and fine tuning processes with the scale of latent consumer GPU compute to open up broader generative AI applications.
Breakpoint 2024
At Breakpoint, Director of the Render Network Foundation, Trevor Harries-Jones, discussed Render Network Foundation Initiatives, progress on RNPs, and the expansion of Render Foundation grants. He concluded by highlighting a key milestone for the Render Network, the upgrading of a majority of all network tokens to the new RENDER SPL token.
Token 2049
At Token 2049, Jules joined a panel of luminaries in the field of DePIN and AI to discuss the future of decentralized computing. During the panel, moderated by Tushar Jain of MultiCoin Capital, Jules introduced his views about how the field of AI moving is towards “inference time compute.” and discussed how the Render Network is building to serve this emerging need by focusing on scaling high performance consumer GPU compute.
Jules noted the similarities between today and when the Render Network got its original product market fit, almost a decade ago, by coordinating latent consumer GPU compute supply to resolve GPU cloud shortages in emerging 3D rendering and spatial computing applications. He discussed how focusing on demand generation and latent underutilized consumer compute has proven to be an effective strategy.
This fall is just the beginning — get involved in community governance!
It’s been an action-packed fall for the Render Network community — but there is still a lot of ongoing work and community governance discussions taking place through year end. Currently, the Render Network Foundation is gathering community feedback for Year 2 Community Rewards in the Render Network Discord. These discussions are informing upcoming RNPs submitted to the Render Network community, so please share feedback and get involved! Finally, work to integrate more tools, render engines, and services is ongoing with new RNPs expected before year end.
We will be sharing broader community highlights, including discussions of new governance initiatives, in coming year end recaps. And, if you haven’t already, please watch ‘765874 — Unification’ which was made possible by a community of decentralized GPUs on the Render Network.
