OpenThinker-Agent-v1-SFT / README.md

EtashGuha

Update README.md

81a9761 verified about 2 hours ago

preview code

raw

history blame contribute delete

6.58 kB

metadata

base_model:
  - Qwen/Qwen3-8B
datasets:
  - OpenThoughts-Agent-v1-SFT
  - OpenThoughts-Agent-v1-RL
library_name: transformers
license: apache-2.0
model-index:
  - name: OpenThinker-Agent-v1
    results: []
pipeline_tag: text-generation
tags:
  - agents
  - terminal
  - code
  - software-engineering

Project | SFT dataset | RL dataset | SFT model | RL model

OpenThinker-Agent-v1-SFT

OpenThoughts-Agent is an open-source effort to curate the best datasets for training agents. Our first release includes datasets, models and our research codebase.

OpenThinker-Agent-v1 is a model trained for agentic tasks such as Terminal-Bench 2.0 and SWE-Bench.

The OpenThinker-Agent-v1 model is post-trained from Qwen/Qwen3-8B. It is SFT-ed on the OpenThoughts-Agent-v1-SFT dataset, then RL-ed on the OpenThoughts-Agent-v1-RL dataset.

This OpenThinker-Agent-v1-SFT model is the model after the SFT stage. For the model after both SFT and RL stages, see OpenThinker-Agent-v1.

Homepage: https://www.openthoughts.ai/blog/agent
Repository: https://github.com/open-thoughts/OpenThoughts-Agent

OpenThinker-Agent-v1 Model Performance

Our OpenThinker-Agent-v1 model is the state-of-the-art model at its scale on agent benchmarks.

Model	Harness	Terminal-Bench 2.0	SWE-Bench Verified	OpenThoughts-TB-Dev
Qwen3-8B	Terminus-2	0.0	0.7	5.7
OpenThinker-Agent-v1	Terminus-2	4.9	15.7	17.3
Qwen3-32B	Terminus-2	1.9	5.7	10.2
Qwen/Qwen3-Coder-30B-A3B-Instruct	OpenHands	10.1	49.2	24.5

Data

We built OpenThinker-Agent-v1 in two stages: supervised fine-tuning, followed by reinforcement learning. Each stage required its own data pipeline – RL tasks (instructions, environments, and verifiers) and SFT traces from strong teacher agents completing tasks.

OpenThoughts-Agent-v1-SFT is an SFT trace dataset containing approximately 15,200 traces drawn from two different data sources we curate:

nl2bash: Simple synthetically generated tasks where the agent has to format shell commands effectively
InferredBugs: A set of bugs in C# and Java collected by Microsoft that we turned into tasks

OpenThoughts-Agent-v1-RL is an RL dataset containing ~720 tasks drawn from the nl2bash verified dataset.

To stabilize training, we built a three-stage filtration pipeline that prunes tasks before they ever hit the learner:

Bad verifiers filter: drop tasks with flaky or excessively slow verifiers.
Environment stability: remove tasks whose containers take too long to build or tear down. Optional difficulty filter: discard tasks that even a strong model (GPT-5 Codex) cannot solve in a single pass.

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 4e-05
train_batch_size: 1
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 16
total_train_batch_size: 16
total_eval_batch_size: 128
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 7.0

Framework versions

Transformers 4.56.0
Pytorch 2.9.0+cu128
Datasets 4.4.1
Tokenizers 0.22.1

Citation

@misc{openthoughts-agent,
  author = {Team, OpenThoughts-Agent},
  month = Dec,
  title = {{OpenThoughts-Agent}},
  howpublished = {https://open-thoughts.ai/agent},
  year = {2025}
}

open-thoughts
/

OpenThinker-Agent-v1-SFT