Collably Network
3.2K posts

Collably Network
@CollablyNetwork
Global Web3 partnership marketplace. Match your project with investors, exchanges, launchpads, service providers, and ecosystem partners.

The 8B model currently training on Agora is 350B tokens in and continuing to converge. The top level metrics and evals look almost exactly like a centralised run. But; - 133 external contributors total bringing 4090's, 5090's, L40S/RTX 6000 and RTX 6000 Pros. These are cards that people actually own - there are no H100, B200's etc. - The max number of nodes the system can support (104) was filled almost immediately. The authorization layer is receiving approximately 100 requests/minute to join. - The total tokens/per second processed moves directly with amount of compute in the swarm, with Agora constantly optimising to make most efficient use of what hardware is present. - MFU is approximately 20%, TPS is 170k tok/s. There are near constant communication failures which Agora is completely absorbing without slowdown. - The system is effectively on auto-pilot, requiring very little intervention from us. Bad nodes are purged immediately before training is affected and new nodes take their place.

















Monday: Darkbloom goes live on OpenRouter. Every OpenRouter user gets a free trial of private inference, powered by idle Macs around the world. And for providers -- next week we're launching an Alpha Rewards Program: earn up to $40 guaranteed just for running a node. Limited time while we're in alpha. Breakdown in the replies. The network is ready for it. Since our Public Alpha two weeks ago: Gemma 4 is now multimodal, SSD caching has made time-to-first-token much faster, and load tests show we can sustain millions of tokens per second -- with more upgrades landing before launch. If you run a node: turn it on. Demand arrives Monday. If you have a Mac: darkbloom(.)dev → one install command and you're earning. The world's sleeping compute is waking up.









