| pretty_name: Intelif v0.1 on the Decision Index | |
| tags: | |
| - intelif | |
| - decision-index | |
| # Intelif v0.1 on the Decision Index 0.2.1 | |
| Results of [`UserMoonlight/intelif-qwen3-4b`](https://huggingface.co/UserMoonlight/intelif-qwen3-4b) at tag `v0.1` on the [Decision Index](https://github.com/apolinario/decision-index) 0.2.1 (kit commit `87d4650b42b377c0291a89c1f1a879f9b31082bf`), run with the packaged engine `intelif.integrations.decision_index:IntelifEngine` from [intelif](https://github.com/SkAndMl/intelif). | |
| | index | raw index | Knowledge & Reasoning | Language Understanding | Retrieval & Classification | Tools & Automation | Arts & Human Taste | median latency | | |
| |---|---|---|---|---|---|---|---| | |
| | **31.77** | 48.62 | 18.3 | 31.1 | 39.6 | 51.0 | 17.8 | 16.5 ms | | |
| All 150,317 requests were answered, none unsupported. One NVIDIA RTX PRO 6000 Blackwell, CUDA 13.0, torch 2.14. | |
| `runs/intelif-qwen3-4b/` holds `scores.json`, `index.json`, `benchmark-summary.json`, `environment.json`, `status.json` and `results.jsonl.gz`. The results were written with `--compact`: they hold each request's id, payload hash, response and latency, but not the benchmark inputs. | |
| Reproduce with the instructions in [`evals/decision_index/`](https://github.com/SkAndMl/intelif/tree/main/evals/decision_index). | |