Created at 8pm, Jan 4
JHKFUZdUCrypto
0
Render whitepaper
rAbQi70NbUr0hW0KT1Zh2whG6mawSaebQOi9yAtSJD8
File Type
PDF
Entry Count
51
Embed. Model
jina_embeddings_v2_base_en
Index Type
hnsw

All about render

The Render Network and AI The growth of artificial intelligence (AI) has generated an unprecedented demand for computational resources. AI applications necessitate vast computational power, outstripping the capacity of traditional CPU-based systems. Through the Render Network SDK, developers will be able to leverage the network's decentralized GPUs for AI compute tasks ranging from NeRF (Neural Reflectance Field) and LightField rendering processes to generative AI tasks.
id: 23e589eed09e31d20e6c8c76eb2b842e - page: 8
Increasingly, 3D artists are introducing AI generated content into their creative workflows, combining hand created digital artwork with generative AI processing. With the integration of AI toolsets like Stable Diffusion on the Render Network, the network supports the increasing convergence between traditional and next generation creative workflows that leverage AI. For example, artists can use artificial intelligence tools to create assets like generative AI textures that are used to render ultra-high resolution immersive 3D worlds on the network. Large-scale art collections using generative AI to vary outputs can also be distributed across the networks nodes, enabling creators to frictionlessly create AI art collections at near unlimited scale. The rise of artificial intelligence requires new forms of digital traceability and asset verification. The Render Networks deep levels of on-chain provenance built into each works Render Graph enable licensing 3D models for AI training, or roy
id: fd9a48ab1d200a6bdbe63b7e98c599d9 - page: 8
The Render Network uses AI technology to accelerate and optimize rendering processes. AI denoising in OctaneRender, the engine used in the Render Network client, has been specifically trained to denoise volumes with further training optimizations possible through distributed computing. Scene AI models surface visibility to get maximum speed while denoising and 8 rendering Out-of-Core Geometry and Emissive Objects. These models are trained on perceptual models of Material, Spectra Irradiance, and Scene Data and enable accelerated rendering for more complex scenes when scene data exceeds VRAM capacity. These models are periodically updated and can be further trained using decentralized GPU nodes.
id: 9cd59f48ca7fa483dc2e3c052e3beb58 - page: 8
The Render Network and Virtual Assets All work on the Render Network is hashed, including the ORBX file an artist uploads to the Render Network, each individual frame from a render job, and an animation produced from the completed frames. The hashing data associated with a render job, ORBX file, or individual frames can be minted on chain and included in NFT metadata. In addition to completed render jobs and frames, all scenes uploaded to the Render Network are hashed with representation of a scenes XML data as well as each of the individual assets contained within a scene. The unique hashed IDs for each scene and its assets can be minted in on-chain metadata, providing additional levels of provenance for blockchain assets.
id: d5325f6d5b06597d155b8532567b5b5c - page: 9
How to Retrieve?
# Search

curl -X POST "https://search.dria.co/hnsw/search" \
-H "x-api-key: <YOUR_API_KEY>" \
-H "Content-Type: application/json" \
-d '{"rerank": true, "top_n": 10, "contract_id": "rAbQi70NbUr0hW0KT1Zh2whG6mawSaebQOi9yAtSJD8", "query": "What is alexanDRIA library?"}'
        
# Query

curl -X POST "https://search.dria.co/hnsw/query" \
-H "x-api-key: <YOUR_API_KEY>" \
-H "Content-Type: application/json" \
-d '{"vector": [0.123, 0.5236], "top_n": 10, "contract_id": "rAbQi70NbUr0hW0KT1Zh2whG6mawSaebQOi9yAtSJD8", "level": 2}'