Hugging Face releases RL-trained adapters for Qwen3.8-27B
Hugging Face trained separate LoRA adapters for Qwen3.8-27B inside the Pi, Hermes Agent, and OpenCode harnesses, then merged them into a reusable super-adapter. The company also released the adapters, GGUFs, RL tasks, training recipe, and safeagents plugin needed to reproduce the workflow. Hugging Face reports that the Pi adapter raised Terminal-Bench 2.1 pass@1 from 39% to 55% while reducing average token use from 2.9K to 1.5K per task, though these benchmark results remain vendor-reported.
The release offers a reproducible way to specialize an open model for a particular coding-agent harness instead of relying solely on a larger or more general-purpose model.
Combining RL on a custom task distribution with your favorite agent harness yields more accurate and efficient models for that target harness and task distribution.
Models are not fully optimized for all harnesses and the tasks you’re doing. Instead, they target general performance. So when you combine a model and harness, you’re leaving performance and efficiency on the table for the stuff you’re actually doing. With RL, you can train a custom harness adapter that squeezes everything out of your model for your harness and task combination.
Take Qwen3.8-27B, its Terminal-Bench 2.1 score with the pi harness is 39% which uses on average 2.9K tokens per task. If we freeze those weights and add a custom RL-trained LoRA adapter we can get 55% pass@1, with average token usage of 1.5K. Most of this is down to the model reasoning less, calling less tools, and performing better first time.
Our earlier work (multi-harness RL, harness adapters) showed that RL on your own task distribution, inside your own harness, gives a more accurate and more efficient model. Those experiments covered data-analysis and code-building tasks. Here we wanted to push this further by exploring whether the result holds at 27B, on harder real-world software engineering tasks, across three harnesses: Pi, OpenCode and Hermes Agent. And most of all, share a fully reusable recipe, so you can reproduce this on your own task, harness, model combinations.
#Hugging Face#Qwen3.8-27B#TRL#AsyncGRPOTrainer#OpenEnv#Harbor#Pi#Hermes Agent#OpenCode#Terminal-Bench 2.1#SWE-bench Pro hard-51#safeagents