Skip to content

Commit

Permalink
Merge pull request #481 from zachhorn/scaling-the-supercloud-updates-2
Browse files Browse the repository at this point in the history
updates
  • Loading branch information
zachhorn authored Jan 9, 2025
2 parents 82d6a9a + 4aa724e commit d89ed93
Showing 1 changed file with 7 additions and 5 deletions.
12 changes: 7 additions & 5 deletions src/content/Blog/scaling-the-supercloud/index.md
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,8 @@ bannerImage: ./scaling-the-supercloud-main.png

*A year of growth and adoption for Akash in 2024 and the vision for 2025 and beyond*

---

This is the decade that will define the next century. Last year, we wrote about two competing forces: one in favor of slowing (or even reversing) progress and another stronger force pushing for accelerating progress. As 2024 unfolded, the world seemed to tip decisively toward acceleration. We see it manifested in open infrastructures, the push for abundant energy, and the belief that broader access to compute benefits all of society. Akash is positioned entirely for a world that demands acceleration.

Overclock Labs, the creators and largest contributors to Akash, has long championed a radically open approach to building permissionless infrastructure. This ethic is rooted in solid fundamentals, which genuinely matter when building for the future. 2024 was the year that tested and ultimately affirmed our open approach to development. From record-breaking network growth to strategic integrations, from expansions into new regions to steadily climbing adoption, every milestone was propelled by the same undercurrent — open collaboration that thrives in public.
Expand Down Expand Up @@ -61,28 +63,28 @@ Throughout 2024, a wave of new entrants joined the Akash ecosystem, each drawn b
In April, [Brev.dev](https://www.nvidia.com/en-us/launchables/) integrated the Akash Supercloud to expand its platform compute offerings by leveraging NVIDIA GPUs from Akash. This integration increased demand for GPUs on Akash while giving Brev users a wider selection of GPUs at a range of price points. Brev and Akash also co-sponsored a booth at NVIDIA GTC in San Jose — and only a few months later, [Brev was acquired by NVIDIA](https://x.com/NaderLikeLadder/status/1813286240093151412).

#### Venice.ai
[Venice.ai](http://Venice.ai) integrated with Akash to access GPUs for inference of the leading open-source AI models. By leveraging the Akash Supercloud, Venice.ai gave users access to the most capable AI models (including Llama 3.1 405B) through a permissionless and privacy-preserving chat application. The integration demonstrated impressive performance metrics, handling hundreds of thousands of user requests and processing billions of tokens in the months following the integration.
[Venice.ai](http://Venice.ai) integrated with Akash to [access GPUs for inference of the leading open-source AI models](https://x.com/akashnet_/status/1838650155987566899). By leveraging the Akash Supercloud, Venice.ai gave users access to the most capable AI models (including Llama 3.1 405B) through a permissionless and privacy-preserving chat application. The integration demonstrated impressive performance metrics, handling hundreds of thousands of user requests and processing billions of tokens in the months following the integration.

#### Passage.io
[Passage.io](http://Passage.io) tapped into the Akash Supercloud to render immersive virtual worlds and significantly reduced compute costs. By leveraging Akash's decentralized marketplace, [Passage reduced cloud computing costs by an average of 50%](https://akash.network/blog/passage-reduces-cloud-spend-by-50-with-akash/) (up to 70% in some cases) compared to traditional providers. This integration enabled Passage to maintain its zero-compromise approach to browser-based virtual experiences while significantly reducing infrastructure costs. This integration further validates the scale and cost-effectiveness of Akash's permissionless model and positions Passage to accelerate platform development and feature expansion for creators worldwide.

#### Prime Intellect
[Prime Intellect](https://www.primeintellect.ai/) expanded its GPU marketplace by incorporating Akash’s wide range of high-performance GPUs, including NVIDIA H100, A100, A6000, and more. Using Akash's decentralized infrastructure, developers can quickly configure and deploy GPU clusters through Prime Intellect's interface. This integration further streamlines access to Akash compute, and brings GPUs on the Akash Supercloud to a broader set of developers.
[Prime Intellect](https://www.primeintellect.ai/) expanded its GPU marketplace by [incorporating Akash’s wide range of high-performance GPUs](https://x.com/akashnet_/status/1807865087161405460), including NVIDIA H100, A100, A6000, and more. Using Akash's decentralized infrastructure, developers can quickly configure and deploy GPU clusters through Prime Intellect's interface. This integration further streamlines access to Akash compute, and brings GPUs on the Akash Supercloud to a broader set of developers.

#### Quai Network
[Quai Network](https://qu.ai/) integrated with Akash Network to enhance its scalable Proof-of-Work blockchain infrastructure. [This integration](https://qu.ai/blog/quai-network-teams-up-with-akash-to-transform-blockchain-computing/) enables Quai miners to access decentralized operations through Akash nodes, strengthening network security and expanding participation. Developers gained streamlined access to Akash's global compute marketplace, reducing barriers to entry for deploying Quai services and dApps. The integration supports Quai's dual-token system while leveraging Akash compute to improve transaction processing speeds and reduce network latency.

#### Witness Chain
[Witness Chain](https://www.witnesschain.com/) integrated with Akash Network to decentralize its Proof of Location infrastructure. The integration enables the deployment of Witness Chain's Watchtower nodes on Akash, enhancing the network's ability to validate real-world attributes like location and connectivity. Witness Chain Watchtowers leverage Akash's global network to perform internet telemetry-based delay measurements and cryptographic pings. This collaboration strengthens the resilience of Witness Chain's observation layer while exploring potential applications for measuring the decentralization of Akash's provider network.
[Witness Chain](https://www.witnesschain.com/) integrated with Akash Network to decentralize its Proof of Location infrastructure. [The integration](https://x.com/akashnet_/status/1859279274915611001) enables the deployment of Witness Chain's Watchtower nodes on Akash, enhancing the network's ability to validate real-world attributes like location and connectivity. Witness Chain Watchtowers leverage Akash's global network to perform internet telemetry-based delay measurements and cryptographic pings. This collaboration strengthens the resilience of Witness Chain's observation layer while exploring potential applications for measuring the decentralization of Akash's provider network.

#### FLock.io
[FLock.io](http://FLock.io) integrated with Akash to enable one-click deployment templates for FLock nodes. This integration will give AI developers direct access to enterprise-grade GPUs like NVIDIA H100s and A100s at competitive pricing, helping to democratize AI development by making advanced machine learning more accessible to developers around the world. By combining FLock's training platform with Akash's compute marketplace, the integration will reduce technical barriers in decentralized AI training and lay the groundwork for expanding decentralized AI capabilities in 2025\.
[FLock.io](http://FLock.io) integrated with Akash to [enable one-click deployment templates for FLock nodes](https://x.com/akashnet_/status/1805635609165287474). This integration will give AI developers direct access to enterprise-grade GPUs like NVIDIA H100s and A100s at competitive pricing, helping to democratize AI development by making advanced machine learning more accessible to developers around the world. By combining FLock's training platform with Akash's compute marketplace, the integration will reduce technical barriers in decentralized AI training and lay the groundwork for expanding decentralized AI capabilities in 2025\.

#### ElizaOS by ai16z
AI Agents are emerging as one of the major themes of 2025\. In December, the [AkashChat API was integrated](https://x.com/akashnet_/status/1869117078998745282) into [ElizaOS by ai16z](https://elizaos.ai/), which allows agent builders to access the leading open-source AI models directly from the Akash Supercloud. Autonomous agents will need access to compute, and a permissionless cloud like Akash is perfectly aligned to meet that demand.

#### Morpheus
[Morpheus](https://mor.org/) is integrating with Akash to provide decentralized computing infrastructure for AI Smart Agents. This integration allows Smart Agents to access compute resources and run AI model inference on Akash's decentralized network, rather than relying on traditional cloud providers. Morpheus has created a compute marketplace that connects directly with Akash, enabling Smart Agents to independently pay for and access the computing power they need. This integration forms a key part of the Morpheus "Smart Agent Stack," where Akash, along with TAO subnets, can host AI models that Smart Agents can use for inference as they process tasks.
[Morpheus](https://mor.org/) is integrating with Akash to [provide decentralized computing infrastructure for AI Smart Agents](https://x.com/akashnet_/status/1857167384400089562). This integration allows Smart Agents to access compute resources and run AI model inference on Akash's decentralized network, rather than relying on traditional cloud providers. Morpheus has created a compute marketplace that connects directly with Akash, enabling Smart Agents to independently pay for and access the computing power they need. This integration forms a key part of the Morpheus "Smart Agent Stack," where Akash, along with TAO subnets, can host AI models that Smart Agents can use for inference as they process tasks.

#### The University of Texas at Austin & the Rochester Institute of Technology
In the past year, Akash has gained significant recognition from the academic community. Most notably, a research team from [The University of Texas at Austin](https://www.utexas.edu/) (UT-Austin), led by [Professor Sriram Vishwanath](https://www.ece.utexas.edu/people/faculty/sriram-vishwanath), is exploring ways to achieve verifiable compute on Akash. This research was complemented by work from the [Rochester Institute of Technology](https://www.rit.edu/) (RIT), which leveraged Akash's permissionless compute capabilities to advance academic research in decentralized computing and permissionless AI. Through these academic collaborations, students and researchers have tested and expanded the potential applications of decentralized compute. The growing adoption across enterprises, startups, and universities has demonstrated Akash's versatility, establishing the network as a leader in permissionless infrastructure.
Expand Down

0 comments on commit d89ed93

Please sign in to comment.