GOSHUNCLE commited on
Commit
b4c2cb5
·
verified ·
1 Parent(s): 2f2e19f

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +4 -9
README.md CHANGED
@@ -1,12 +1,7 @@
1
  ---
2
  license: other
3
- language:
4
- - zh
5
- - en
6
- tags:
7
- - text-classification
8
- - llm-router
9
- - cost-optimization
10
  ---
11
 
12
  # llm-cost-router — Stage 2
@@ -15,12 +10,12 @@ tags:
15
 
16
  ## 中文
17
  本模型為「AI 路由(LLM Cost-Aware Router)」專案的任務分類模型之一,
18
- 以 microsoft/mdeberta-v3-base 微調,判斷企業員工送給 LLM 的請求該如何路由,
19
  以節省昂貴模型的呼叫成本。訓練資料為電子製造業情境的合成資料 + 少量人工校正。
20
 
21
  ## English
22
  One of the classifiers of the *LLM Cost-Aware Router* project, fine-tuned from
23
- microsoft/mdeberta-v3-base to route enterprise LLM requests and cut expensive-model cost.
24
  Trained on synthetic electronics-manufacturing data with light human correction.
25
 
26
  ## 限制 / Limitations
 
1
  ---
2
  license: other
3
+ language: [zh, en]
4
+ tags: [text-classification, llm-router, cost-optimization]
 
 
 
 
 
5
  ---
6
 
7
  # llm-cost-router — Stage 2
 
10
 
11
  ## 中文
12
  本模型為「AI 路由(LLM Cost-Aware Router)」專案的任務分類模型之一,
13
+ 以 xlm-roberta-base 微調,判斷企業員工送給 LLM 的請求該如何路由,
14
  以節省昂貴模型的呼叫成本。訓練資料為電子製造業情境的合成資料 + 少量人工校正。
15
 
16
  ## English
17
  One of the classifiers of the *LLM Cost-Aware Router* project, fine-tuned from
18
+ xlm-roberta-base to route enterprise LLM requests and cut expensive-model cost.
19
  Trained on synthetic electronics-manufacturing data with light human correction.
20
 
21
  ## 限制 / Limitations