3 Commits
v1.0 ... v1.0.1

Author SHA1 Message Date
4paradigm
a0d9bdf96e fix README.md
All checks were successful
Docker Build and Push / docker (push) Successful in 1m5s
2026-08-05 14:13:43 +08:00
4paradigm
750992c82d fix tokenizer
All checks were successful
Docker Build and Push / docker (push) Successful in 1m5s
2026-07-22 14:40:10 +08:00
4paradigm
500f2bba0d fix tokenizer 2026-07-22 14:38:23 +08:00
2 changed files with 24 additions and 6 deletions

View File

@@ -7,7 +7,7 @@ on:
jobs:
docker:
runs-on: amd64-ubuntu-24.04
runs-on: arm64-ubuntu-24.04
steps:
- name: Clone repository

View File

@@ -6,10 +6,10 @@
部分模型的 `tokenizer_config.json` 存在以下问题,导致 vLLM 服务启动失败(错误码 `TOKENIZER_FAILED`
| 错误 | 原因 | 受影响模型数 |
|---|---|---|
| `ValueError: Tokenizer class TokenizersBackend does not exist` | `tokenizer_class` 不是 transformers 合法类名 | 10 |
| `AttributeError: 'list' object has no attribute 'keys'` | `extra_special_tokens` 为 list 格式transformers 要求 dict | 15 |
| 错误 | 原因 |
|---|---|
| `ValueError: Tokenizer class TokenizersBackend does not exist` | `tokenizer_class` 不是 transformers 合法类名 |
| `AttributeError: 'list' object has no attribute 'keys'` | `extra_special_tokens` 为 list 格式transformers 要求 dict |
详见 `tokenizer_error_classification_summary.md`
@@ -57,7 +57,25 @@ docker run -dit --name <container_name> \
--trust-remote-code --max-model-len 2048 --enforce-eager -tp 1
```
## 构建
## 测试
如果发送请求时出现以下报错:
```
ValueError: default chat template is no longer allowed, so you must provide a chat template if the tokenizer does not define one.
```
请使用 completion 接口测试,而非 chat completion
```bash
curl http://localhost:8000/v1/completions \
-H "Content-Type: application/json" \
-d '{
"model": "llm",
"prompt": "Hello, how are you?",
"max_tokens": 100
}'
```
```bash
docker build -t vllm-ascend-fix-tokenizer:latest .