Update README.md

README.md

@@ -38,6 +38,46 @@ benchmark documents.
 | Primary language | Arabic |
 | Secondary language | English |
 
+## Arabic tokenizer benchmark
+
+SamAI-Tokenizer was evaluated on four complementary Arabic benchmarks. Lower
+tokens per word indicates more efficient tokenization.
+
+![SamAI tokenizer benchmark results](assets/SamAI-Tokenizer-Benchmark.png)
+
+| Benchmark | SamAI tokens/word | Best tokenizer | Best tokens/word | SamAI rank |
+| --- | ---: | --- | ---: | ---: |
+| Arabic Dialects | **1.3043** | ALLaM-7B | 1.2812 | **2** |
+| ArabicMMLU | **2.4838** | ALLaM-7B | 1.9943 | **5** |
+| MASSIVE Arabic | **1.2549** | ALLaM-7B | 1.1939 | **2** |
+| MLQA Arabic | **1.4397** | **SamAI-Tokenizer** | **1.4397** | **1** |
+
+Across the equal-weight macro average of the four benchmarks, SamAI-Tokenizer
+ranked second among 15 successfully tested tokenizers at **1.6207 tokens per
+word**.
+
+| Rank | Tokenizer | Mean tokens/word | Extra tokens versus SamAI |
+| ---: | --- | ---: | ---: |
+| 1 | ALLaM-7B | 1.5021 | -4.86% |
+| **2** | **SamAI-Tokenizer** | **1.6207** | **baseline** |
+| 3 | Falcon-H1-7B | 1.8880 | +19.30% |
+| 4 | Mistral-Small-3.1 | 1.8929 | +20.71% |
+| 5 | Qwen3.8-Flash-Next | 1.9039 | +21.94% |
+| 6 | Qwen3.5-4B | 1.9039 | +21.94% |
+| 7 | tiktoken-o200k | 1.9919 | +26.37% |
+| 8 | GPT-OSS-20B | 1.9919 | +26.37% |
+| 9 | Fanar-1 | 2.1570 | +36.81% |
+| 10 | DeepSeek-V3 | 2.2774 | +46.25% |
+| 11 | Qwen2.5-7B | 2.3740 | +50.25% |
+| 12 | Qwen3-8B | 2.3740 | +50.25% |
+| 13 | GLM-5.3-Flash | 2.5225 | +60.21% |
+| 14 | Kimi-K2.6 | 2.8693 | +84.13% |
+| 15 | tiktoken-cl100k | 4.0800 | +164.63% |
+
+Complete aggregate results are available in [`results/`](results/). Row-level
+outputs are intentionally excluded because they reproduce third-party text.
+
+
 ## Installation
 
 ```bash
@@ -107,45 +147,6 @@ dialects, Arabizi, English, code, mathematics, and Unicode edge cases.
 See [`reports/validation.json`](reports/validation.json) and
 [`reports/stress.json`](reports/stress.json).
 
-## Arabic tokenizer benchmark
-
-SamAI-Tokenizer was evaluated on four complementary Arabic benchmarks. Lower
-tokens per word indicates more efficient tokenization.
-
-![SamAI tokenizer benchmark results](assets/SamAI-Tokenizer-Benchmark.png)
-
-| Benchmark | SamAI tokens/word | Best tokenizer | Best tokens/word | SamAI rank |
-| --- | ---: | --- | ---: | ---: |
-| Arabic Dialects | **1.3043** | ALLaM-7B | 1.2812 | **2** |
-| ArabicMMLU | **2.4838** | ALLaM-7B | 1.9943 | **5** |
-| MASSIVE Arabic | **1.2549** | ALLaM-7B | 1.1939 | **2** |
-| MLQA Arabic | **1.4397** | **SamAI-Tokenizer** | **1.4397** | **1** |
-
-Across the equal-weight macro average of the four benchmarks, SamAI-Tokenizer
-ranked second among 15 successfully tested tokenizers at **1.6207 tokens per
-word**.
-
-| Rank | Tokenizer | Mean tokens/word | Extra tokens versus SamAI |
-| ---: | --- | ---: | ---: |
-| 1 | ALLaM-7B | 1.5021 | -4.86% |
-| **2** | **SamAI-Tokenizer** | **1.6207** | **baseline** |
-| 3 | Falcon-H1-7B | 1.8880 | +19.30% |
-| 4 | Mistral-Small-3.1 | 1.8929 | +20.71% |
-| 5 | Qwen3.8-Flash-Next | 1.9039 | +21.94% |
-| 6 | Qwen3.5-4B | 1.9039 | +21.94% |
-| 7 | tiktoken-o200k | 1.9919 | +26.37% |
-| 8 | GPT-OSS-20B | 1.9919 | +26.37% |
-| 9 | Fanar-1 | 2.1570 | +36.81% |
-| 10 | DeepSeek-V3 | 2.2774 | +46.25% |
-| 11 | Qwen2.5-7B | 2.3740 | +50.25% |
-| 12 | Qwen3-8B | 2.3740 | +50.25% |
-| 13 | GLM-5.3-Flash | 2.5225 | +60.21% |
-| 14 | Kimi-K2.6 | 2.8693 | +84.13% |
-| 15 | tiktoken-cl100k | 4.0800 | +164.63% |
-
-Complete aggregate results are available in [`results/`](results/). Row-level
-outputs are intentionally excluded because they reproduce third-party text.
-
 ## Benchmark methodology
 
 - All available examples were used, followed by exact-text deduplication within