Apriel-1.5-OpenReasoner: RL Post-Training for General-Purpose and Efficient Reasoning

TL;DR

Apriel-1.5-OpenReasoner is a reinforcement learning post-training method for large language models that enhances multi-domain reasoning efficiency and accuracy while reducing inference costs and trace lengths.

Contribution

It introduces an adaptive domain sampling and difficulty-aware length penalty to improve reasoning across diverse tasks without additional training overhead.

Findings

Outperforms Apriel-Base on multiple reasoning benchmarks.

Produces 30-50% shorter reasoning traces.

Generalizes to longer token outputs at inference.

Abstract

Building general-purpose reasoning models using reinforcement learning with verifiable rewards (RLVR) across diverse domains has been widely adopted by frontier open-weight models. However, their training recipes and domain mixtures are often not disclosed. Joint optimization across domains poses significant challenges: domains vary widely in rollout length, problem difficulty and sample efficiency. Further, models with long chain-of-thought traces increase inference cost and latency, making efficiency critical for practical deployment. We present Apriel-1.5-OpenReasoner, trained with a fully reproducible multi-domain RL post-training recipe on Apriel-Base, a 15B-parameter open-weight LLM, across five domains using public datasets: mathematics, code generation, instruction following, logical puzzles and function calling. We introduce an adaptive domain sampling mechanism that preserves target domain ratios despite heterogeneous rollout dynamics, and a difficulty-aware extension of the standard length penalty that, with no additional training overhead, encourages longer reasoning for difficult problems and shorter traces for easy ones. Trained with a strict 16K-token output budget, Apriel-1.5-OpenReasoner generalizes to 32K tokens at inference and improves over Apriel-Base on AIME 2025, GPQA, MMLU-Pro, and LiveCodeBench while producing 30-50% shorter reasoning traces. It matches strong open-weight models of similar size at lower token cost, thereby pushing the Pareto frontier of accuracy versus token budget.