Quasar
Open long-context foundation models trained on decentralized compute
Quasar runs long-context LLM competitions — miners build models for infinite-context tasks.
The simplest explanation
Like a reading comprehension competition for AI, but the documents can be book-length.
Everything you need to understand Quasar
No jargon. No whitepapers. Just the facts.
The Problem
Standard transformers scale attention quadratically, making very long context windows expensive, and the leading long-context models are closed, gated, and trained on centralized GPU clusters out of reach for most builders.
What Quasar Does
Quasar runs long-context LLM competitions — miners build models for infinite-context tasks.
The Edge
Quasar uses linear continuous-time attention aiming to handle millions of tokens at a fraction of the compute, and trains on Bittensor's distributed miner network rather than a central cluster. Every model ships with full open weights under Apache 2.0, no waitlist or gating.
What's the closest mainstream equivalent?
The best way to understand a new technology is to compare it to something familiar.
Quasar uses linear continuous-time attention aiming to handle millions of tokens at a fraction of the compute, and trains on Bittensor's distributed miner network rather than a central cluster. Every model ships with full open weights under Apache 2.0, no waitlist or gating.
How does Bittensor make this possible?
Quasaruses Bittensor's incentive layer to build something no single company could run alone.
Miners compete
Participants (called miners) on the Quasar subnet compete to produce the best outputs for science tasks. Anyone with the right hardware can join.
Validators score the work
Validators continuously evaluate miner outputs against objective benchmarks. The best performers rise, the worst are replaced. There's no human committee — the protocol decides.
Rewards flow to the best
Miners are paid in the Quasar alpha token in proportion to how good their work is. This creates a continuous competitive pressure that drives quality up and cost down — structurally, not just as a promise.
The whole network benefits
Because every participant is aligned toward the same goal — producing the best science results — the network improves continuously without requiring a central team to manage it.
Frequently Asked Questions about Quasar
What is Quasar on Bittensor?
Quasar is Subnet 24 (SN24) on the Bittensor network — a decentralized AI protocol built on the TAO blockchain. Quasar runs long-context LLM competitions — miners build models for infinite-context tasks.
What problem does Quasar solve?
Standard transformers scale attention quadratically, making very long context windows expensive, and the leading long-context models are closed, gated, and trained on centralized GPU clusters out of reach for most builders.
How is Quasar different from Closed long-context frontier models (Google Gemini, OpenAI GPT long-context)?
Quasar uses linear continuous-time attention aiming to handle millions of tokens at a fraction of the compute, and trains on Bittensor's distributed miner network rather than a central cluster. Every model ships with full open weights under Apache 2.0, no waitlist or gating.
What is the Quasar token?
The Quasar subnet has its own alpha token on Bittensor's dTAO system. It trades in the Bittensor liquidity pool and its price reflects market demand for the subnet's services.
Is Quasar a good investment?
AlphaGap tracks Quasar using its aGap score — a composite of development activity, token flow, and social signals. This page is for informational purposes only and is not financial advice. Always do your own research before making any investment decisions.
Want real-time intelligence
on Quasar?
AlphaGap tracks signals, whale flows, developer commits, and the aGap score for every Bittensor subnet — updated continuously. Find the alpha gap before everyone else.
Free preview · Pro from $29/mo · Premium from $49/mo
