datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
R2E-Gym-LiteR2E-Gym-V1R2E-Gym-SubsetSWE-Bench-VerifiedR2E-Gym-Subset-Verified
R2E-Gym-Subset-Verified
Gold-patch-validated subset of
R2E-Gym/R2E-Gym-Subset
(paper). The train split contains
4,522 / 4,578 rows (98.78%) verified scoreable end-to-end: apply the gold patch, run the
upstream /testbed/run_tests.sh baked into the row's image, check the parsed outcomes against
expected_output_json.
Changes vs upstream
Validation-only subset — our passes, run in fresh sandboxes per row: one full pass at
concurrency 200, then a 10× retry pass over… See the full description on the dataset page: https://huggingface.co/datasets/PrimeIntellect/R2E-Gym-Subset-Verified.SWE-Bench-LiteR2EGym-SFT-TrajectoriesR2E-Gym-Full
R2E-Gym Subset Filtered for MAGRPO
Filtered subset of R2E-Gym optimized for 2-agent MAGRPO training with 7B models.
Dataset Statistics
Total instances: 167
Format: Issue description + Oracle files in prompt
Optimized for: 2-agent collaboration, 7B models
Filtering Criteria (SWE-bench Lite Style)
Problem statement: >40 words (up to 500 for context window)
Must have non-empty oracle patch (non-test file changes)
File count: Exactly 1 oracle file (single-file… See the full description on the dataset page: https://huggingface.co/datasets/ryankamiri/R2E-Gym-Full.R2E-Gym-CollabR2EGym-Verifier-Trajectoriesrl__24GPU_base__mix_h2_language_balanced__r2egym-nl2bash-stackR2E-Gym-Subset-OracleR2E-Gym-SubsetR2EGym-TestingAgent-SFT-Trajectoriesrl__24GPU_base__swe_rebench_patched_oracle__r2egym-nl2bash-stackR2E-TestgenAgent-PatchesR2E-Gym-Lite-Truncate-7B-FixedR2E-Gym-Subset-contextR2EGym-VerifierTrajectories-PatchOnlyrl__24GPU_base__code-contests-noblock__r2egym-nl2bash-stackrl__24GPU_shaped__swe_rebench_patched_oracle__r2egym-nl2bash-stackR2E-Gym-Lite-RFT-no-thinka3-rl-DCAgent_r2egym-patched-full-oracleR2E-Gym
R2E-Gym: OpenCode, Codex, and Claude Code environments
This repository contains three environment variants of
R2E-Gym/R2E-Gym-Subset.
Each variant contains the same 4,578 tasks in five Parquet shards, with the
original 14-column schema and task order. Only docker_image is replaced with
the corresponding public image containing the selected agent.
Configuration
Files
Docker image prefix
opencode
opencode/data/*.parquet
docker.io/loongsage/r2e-gym:opencode_
codex… See the full description on the dataset page: https://huggingface.co/datasets/loongsage/R2E-Gym.R2E-Gym-Lite-Truncate-7BR2E-Gym-Lite-RFTr2egym-patched-full-oracle-qwen3.5-122b-131k-opencode-literal-rescue-traces
Agent trace dataset
Decoding the literal token IDs
The prompt_token_ids / completion_token_ids / logprobs columns are the
verbatim tokens the serving engine emitted, stored PER AGENT STEP as a
list-of-lists (one inner list per turn). To turn them back into text you MUST
use the exact tokenizer the model was served with — a generic same-family
tokenizer will decode word tokens to garbage.
Served model / tokenizer source: Qwen/Qwen3.5-122B-A10B-FP8
from transformers… See the full description on the dataset page: https://huggingface.co/datasets/open-athena/r2egym-patched-full-oracle-qwen3.5-122b-131k-opencode-literal-rescue-traces.r2egym_gpt5mini_1500ie1_gpt_long_r2egym_sandboxes_4x_glm_4.7_traces_jupiterR2E-Gym-Yiming-Combined-1p6pct