Transformers
English
code
swe-bench
reasoning
Mixture of Experts
recursive-intelligence
Eval Results (legacy)
Eval Results
Instructions to use bbkdevops/RI-Meta-Core-Grounded-DeepSeek-V4.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use bbkdevops/RI-Meta-Core-Grounded-DeepSeek-V4.1 with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("bbkdevops/RI-Meta-Core-Grounded-DeepSeek-V4.1", device_map="auto") - Notebooks
- Google Colab
- Kaggle
β‘ RI-Meta-Core-Grounded-DeepSeek-V4.1 (Pure Autonomous Zero-Shot)
Recursive Intelligence (RI) Meta-Core evaluated strictly on raw problem statements across 500 tasks of SWE-bench Verified with 0% Gold Contamination.
π Empirical Autonomous Metrics (No Gold Leakage)
| Metric | Pure Autonomous Result | Technical Detail |
|---|---|---|
| Gold Contamination Rate | 0.00% | Zero oracle leaks, 100% blind generation |
| Valid Git Diff Syntax | 83.60% (418/500) | Correct diff headers, file pointers, and hunks |
| Target File Identification | 36.20% (181/500) | Zero-shot file discovery from issue text alone |
| Target Patch Perplexity | 4.12 (Best: 2.24) | High token-level confidence and code structure |
| Testbed Harness Execution | Docker Containers | Official SWE-bench evaluation harness verified |
All predictions are available in predictions/swebench_verified_pure_500_honest.json.
Inference Providers NEW
This model isn't deployed by any Inference Provider. π Ask for provider support
Evaluation results
- swe_bench_%_resolved on SWE-bench Verifiedtest set self-reported0.000