Research
Experiments in decentralized post-training. Read the methods, inspect the results, and build on released checkpoints.
Experiments & results.
Each publication records its own model, method and limitations.
- Read the article
A Market Mechanism for Rollout Selection in Group-Relative RL
A 4B model trained for 12,852 updates on rollouts supplied entirely by independent operators: +31.1pp math pass@1 and +20.1pp code over Qwen3-4B-Base on exact-content-disjoint subsets, with transfer to AMC23, AIME, HumanEval+ and MBPP+. Single run, not replicated, and no matched centralized control — the report states what it could not measure.
- Read the article
Eight rollouts are not eight votes
A production window, the complete 0–8 frontier, 1,104 production trajectories in a frozen ten-window sample, and a brick-by-brick protocol for testing behavioral diversity against causal training utility.
84 / 138
groups at 2 / 8 reward-positive
- Read the article
Reliquary: a market for the learning frontier
How Subnet 81 turns prompt selection into competition among operators: miners bet compute on prompts at the policy's learning frontier, replacing DAPO's reactive filter with an ex-ante commitment — +14pp pass@1 over vanilla GRPO at identical step count (Qwen3-4B-Instruct, 300 steps, held-out math, 0.47 → 0.61, single run, not replicated).
- Read the article
Reliquary 2B: training done
Two environments, one system, two outcomes. Code gained accuracy while getting shorter; math learned to terminate. What the constraint decided, and what a post-trained checkpoint cannot tell us.
- Read the article
Paying for what we can actually consume
Emission moves from a fixed amount distributed by rank to a price per unit of admissible data, and what the trainer cannot consume burns. The sensor, the controller, the task registry, the payment floor, and what we will measure.
From independent GPUs to a released model.
The Reliquary-4B run: how it worked, what improved, and what remains to measure.
RESEARCH PAPERTECHNICAL REPORT / SEPTEMBER 2026
A Market Mechanism for Rollout Selection in Group-Relative RL
Independent operators supplied the rollouts. Reliquary trained Qwen3-4B-Base on math and code, then released the weights and evaluation results. Read the full method, figures and limitations; this is a single reported run, not a savings guarantee.
- Model
- Reliquary-4B
- Report published
- 2026-09-21
- Evaluated release
- 01d004d8
Take the next step.
Current workspace and API guides live in the app. Research publications remain here.
- 01 / guide · getting started
Make your first request.
Read the public inventory without a wallet or API key.
- 02 / guide · platform
Prepare a training-data batch.
Choose compatible inputs, check availability, and understand quotes and delivery.
- 03 / guide · network
Read the network evidence.
Keep selection, training, publication and evaluation distinct.
- 04 / guide · operators
Contribute compute.
Find maintained miner setup guides and runtime requirements.
Build on the work.
Cite the training report for the published experiment. Use the maintained documentation for today’s network and API behavior.
@techreport{reliquary2026rollout,
author = {Fanucci, Romain and El, Malik},
title = {A Market Mechanism for Rollout Selection in Group-Relative RL},
institution = {Reliquary},
year = {2026},
month = {9},
type = {Technical report},
note = {Not peer reviewed},
url = {https://reliqua.ai/research/rollout-selection-mechanism/paper}
}