Update README.md

README.md

@@ -30,27 +30,11 @@ base_model:
 
 **Reasoning-first · Zero benchmark contamination · Trained on a free Colab T4**
 
-
-
 ![License](https://img.shields.io/badge/license-CC--BY--4.0-blue)
-
-
-
-
 ![Base](https://img.shields.io/badge/base-Qwen3--1.7B-orange)
-
-
-
-
 ![Method](https://img.shields.io/badge/method-GRPO-purple)
-
-
-
-
 ![Params](https://img.shields.io/badge/params-1.7B-green)
 
-
-
 </div>
 
 ---
@@ -114,6 +98,18 @@ These are the results of the benchmarks for *Atomight-V2.5-1.7B*, evaluated with
 
 ---
 
+## IIfSLM Benchmark Results
+
+[IIfSLM](https://huggingface.co/datasets/NovatasticRoScript/IIfSLM-v1) (Intelligence Index for Small Language Models) is an open, contamination-resistant benchmark suite for the 0.5B–4B range — built for the whole small-model community to evaluate against, not exclusive to this model.
+
+| Domain | Score | Details |
+|---|---:|---|
+| gsm8krefn | **86.96%** | 260/299 correct · greedy decoding · max_new_tokens=900 |
+
+More domains (HumanEval, ARC-Challenge) are in progress. See the [full IIfSLM dataset and methodology notes](https://huggingface.co/datasets/NovatasticRoScript/IIfSLM-v1) — and feel free to run your own model against it too.
+
+---
+
 ## How it compares with other small language models (we recommend verifying it, as the other data from other models came from a third-party sources) 
 
 Scores below for other models are drawn from their respective model cards / technical reports, not re-run by us. Provided for context only — evaluation harnesses and prompt formats differ across labs, so treat this as directional rather than exact.
@@ -178,4 +174,5 @@ Training method — this model was trained using GRPO (Group Relative Policy Opt
   journal={arXiv preprint arXiv:2402.03300},
   year={2024}
 }
-```
+```
+