Update README.md

README.md

@@ -1,62 +1,46 @@
 ---
-license: apache-2.0
 language:
-- bn
 - en
-base_model: Qwen/Qwen3.6-27B
 tags:
-- qwen3_5
-- fine-tuned
+- model
+- leaderboard
 model-index:
-- name: DropLychee-1.2
+- name: আপনার-মডেলের-নাম
   results:
   - task:
-      type: text-generation
-      name: GPQA Diamond (zero-shot)
+      type: multiple-choice  # টাস্ক টাইপ (যেমন: text-generation, multiple-choice)
+      name: GPQA
     dataset:
-      name: GPQA Diamond
       type: Idavidrein/gpqa
+      name: default          # ডেটাসেটের সাবসেট (বেশিরভাগ ক্ষেত্রে 'default' বসালে চলে)
     metrics:
-    - type: acc_norm
-      value: 42.0
-      name: accuracy (normalized)
+      - type: accuracy       # মেট্রিক্স টাইপ (accuracy, exact_match, pass@1 ইত্যাদি)
+        value: 87.8
   - task:
-      type: text-generation
-      name: MMLU-Pro (5-shot)
-    dataset:
+      type: multiple-choice
       name: MMLU-Pro
+    dataset:
       type: TIGER-Lab/MMLU-Pro
+      name: default
     metrics:
-    - type: exact_match
-      value: 64.0
-      name: exact match
----
-
-# DropLychee 1.2
-
-Fine-tuned model based on Qwen3.6-27B (LoRA fine-tune, 500 steps).
-
-## Evaluation
-
-Evaluated using [lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness) on 2026-07-12.
-
-| Benchmark | Score |
-|---|---|
-| GPQA Diamond (0-shot) | 42.0% (acc_norm) |
-| MMLU-Pro (5-shot) | 64.0% (exact_match) |
-
-Total evaluation time: ~2.3 hours on RTX PRO 6000 Blackwell (96GB).
-
-## Usage
-
-\`\`\`python
-from unsloth import FastLanguageModel
-
-model, tokenizer = FastLanguageModel.from_pretrained(
-    model_name="droplychee/DropLychee-1.2",
-    max_seq_length=2048,
-    dtype=None,
-    load_in_4bit=False,
-)
-FastLanguageModel.for_inference(model)
-\`\`\`
+      - type: accuracy
+        value: 86.2
+  - task:
+      type: text-generation   # কোডিং বেঞ্চমার্কের জন্য সাধারণত text-generation বা code-generation হয়
+      name: SWE Bench Pro
+    dataset:
+      type: ScaleAI/SWE-bench_Pro
+      name: default
+    metrics:
+      - type: resolved        # SWE-bench এ সাধারণত resolved ব্যবহার করা হয়
+        value: 53.5
+  - task:
+      type: multiple-choice
+      name: MathArena AIME 2026
+    dataset:
+      type: MathArena/aime_2026
+      name: default
+    metrics:
+      - type: accuracy
+        value: 94.1
+---