Upgrade DeepSeek V2 to DeepSeek-V4 tokenizer

- Add DeepSeek-V4 Flash ByteLevel BPE tokenizer (128K vocab, 127K merges)
- Remove DeepSeek V2 tokenizer (deepseek-llm)
- Update registry, tests, READMEs

Author: neluca

README.md

@@ -60,7 +60,7 @@ tok.decode(ids)  # → 'hello world'
 import tinybpe
 
 tinybpe.list_models()
-# ['cl100k_base', 'deepseek-llm', 'minicpm', 'o200k_base',
+# ['cl100k_base', 'deepseek-v4', 'minicpm', 'o200k_base',
 #  'p50k_base', 'phi2', 'qwen35', 'r50k_base']
 ```
 
@@ -74,7 +74,7 @@ tinybpe.list_models()
 | `r50k_base` | GPT-2 | 50,256 |
 | `qwen35` | Qwen3.5 (0.8B-35B) | 247,843 |
 | `phi2` | Microsoft Phi-2 | 50,257 |
-| `deepseek-llm` | DeepSeek V2 (7B-Chat) | 100,013 |
+| `deepseek-v4` | DeepSeek-V4 Flash | 127,997 |
 | `minicpm5` | MiniCPM5-1B (ByteLevel BPE) | 130,050 |
 
 ### Training

README_zh.md

@@ -60,7 +60,7 @@ tok.decode(ids)  # → 'hello world'
 import tinybpe
 
 tinybpe.list_models()
-# ['cl100k_base', 'deepseek-llm', 'minicpm', 'o200k_base',
+# ['cl100k_base', 'deepseek-v4', 'minicpm', 'o200k_base',
 #  'p50k_base', 'phi2', 'qwen35', 'r50k_base']
 ```
 
@@ -74,7 +74,7 @@ tinybpe.list_models()
 | `r50k_base` | GPT-2 | 50,256 |
 | `qwen35` | Qwen3.5（0.8B-35B） | 247,843 |
 | `phi2` | Microsoft Phi-2 | 50,257 |
-| `deepseek-llm` | DeepSeek V2（7B-Chat） | 100,013 |
+| `deepseek-v4` | DeepSeek-V4 Flash | 127,997 |
 | `minicpm5` | MiniCPM5-1B（ByteLevel BPE） | 130,050 |
 
 ### 训练分词器

models/README.md

@@ -32,7 +32,7 @@ ids = tok.encode("hello world")
 |---|---|---|---|---|
 | `qwen35` | Qwen3.5 (0.8B-35B) | 247,843 | GPT-2 regex | `Qwen/Qwen3.5-0.8B` |
 | `phi2` | Microsoft Phi-2 | 50,257 | GPT-2 regex | `microsoft/phi-2` |
-| `deepseek-llm` | DeepSeek V2 (7B-Chat) | 100,013 | None (raw) | `deepseek-ai/deepseek-llm-7b-chat` |
+| `deepseek-v4` | DeepSeek-V4 Flash | 127,997 | None (raw) | `deepseek-ai/DeepSeek-V4-Flash` |
 
 ## Model Format