Opening article
Turning Database Grading into RL Rewards: Post-Training Agents with ThinkingBox
Turning Database Grading into RL Rewards: Post-Training Agents with ThinkingBox — announced via Hugging Face on October 22, 2026. We then used the same kind of executable signal as a reward on training tasks outside ThinkingBox-Bench. Each of its 507 tasks in ThinkingBox-Bench (our benchmark dataset) hands an agent a real backend and a set of tools, lets it work, then checks what really changed in the database. So we constructed a training pool from ThinkingBox-format tasks outside the benchmark and used their executable outcomes as rewards.
Model Size
27B
Dense ParametersBenchmark Metric
51.70%
Reported Accuracy / GainDeployment Platform
Hugging Face
Cloud InfrastructureThe Core Development
This is the follow up to our first blog post, The Agent Said It Was Done. The Database Disagreed. ThinkingBox-Training is now available on GitHub .
We built ThinkingBox to evaluate an agent's work by comparing the final database state with the expected outcome. We then used the same kind of executable signal as a reward on training tasks outside ThinkingBox-Bench. For Qwen3.8-27B, average success rose from 51.70% to 60.78%, while tasks completed correctly on all twenty attempts rose from 38 to 107 with 50 GRPO steps.
Figure 1. ThinkingBox turns executable checks of terminal database state into rewards for post-training agents on tasks outside ThinkingBox-Bench.
How It Works & Technical Architecture
For the reference run, we ran full-parameter RL fine-tuned Qwen3.8-27B on 187 tasks using token-mean GRPO. Each update generated 18 groups of 8 rollouts, or 144 episodes, across 24 H100s. Training ran synchronously with parameter and optimizer offload; AdamW used a 1e-6 learning rate and zero weight decay. We ran the training for 50 steps.
Deployment & Availability
So we constructed a training pool from ThinkingBox-format tasks outside the benchmark and used their executable outcomes as rewards. The catch was choosing which tasks are suitable for agent training: most tasks are either already solved or hopeless for the model in front of you, and both teach it nothing.
Strategic Perspective & Ecosystem Impact
Each of its 507 tasks in ThinkingBox-Bench (our benchmark dataset) hands an agent a real backend and a set of tools, lets it work, then checks what really changed in the database. In the first post , we caught a model closing a customer's ticket as resolved while the carrier exception behind it was still open. Nine competent-looking tool calls, one wrong database row.
Then we noticed what we had built. For most tasks, an executable check can return 1 or 0 directly from the final database state. That is not just a grader. This can be a reward function. +
Difficulty is policy-relative: a task is hard for the current model only until that model learns it.
Why it matters
Turning Database Grading into RL Rewards: Post-Training Agents with ThinkingBox — announced via Hugging Face on October 22, 2026 Reported by Hugging Face on October 22, 2026.