Commit Graph

14 Commits

Author SHA1 Message Date
deep-learning-dynamo 16f60dbef9 reducing duplication of *EmbeddingStoreIT 2023-11-18 16:23:29 +01:00
deep-learning-dynamo 21dfc8b317 released 0.24.0 2023-11-12 18:58:31 +01:00
Simon Verhoeven d0399d023b
fix: remove unused imports (#209) 2023-10-09 09:48:00 +02:00
deep-learning-dynamo 315eab8641 released 0.23.0 2023-09-29 14:27:51 +02:00
jingzhou 0b11631691
Add Qwen open sourced 7-billion-parameter and 14-billion-parameter models (#194) 2023-09-29 09:23:34 +02:00
jiangsier-xyz 8851de2e4d
Match the new model response design for qwen (#175)
Include the TokenUsage and FinishReason information in the Response from
qwen.
2023-09-21 11:57:44 +02:00
LangChain4j 21f744868b
Return Response<T> from models (#159) 2023-09-15 14:52:22 +02:00
jiangsier-xyz f2bb6f992e
Refactor the implementation for Qwen series models using the new DashScope SDK APIs. (#155)
The design of the Dashscope SDK is evolving towards OpenAI, offering new
fields and specifications. Utilize these latest features to refactor the
implementation of the Qwen models.
2023-09-03 13:21:15 +02:00
jiangsier-xyz 80c3880062
Wrap a method to return original LLM result (#147)
Wrap a method to return the original LLM result, allowing subclasses to
have a chance to obtain more information such as usage (input/output
tokens, finish reason...).
2023-08-30 21:14:26 +02:00
deep-learning-dynamo c1cc5be1c7 released 0.22.0 2023-08-29 19:21:56 +02:00
LangChain4j 6f5845ff4e
Text segment improvements (#134) 2023-08-28 20:35:57 +02:00
deep-learning-dynamo db1f236ed2 released 0.21.0 2023-08-19 15:57:39 +02:00
deep-learning-dynamo d013fb0aaa removed unnecessary code 2023-08-18 21:01:25 +02:00
jiangsier-xyz d908f5158a
Integrate the Qwen series models via dashscope-sdk. (#99)
Qwen series models are provided by Alibaba Cloud. They are much better
in Asia languages then other LLMs.

DashScope is a model service platform. Qwen models are its primary
supported models. But it also supports other series like LLaMA2, Dolly,
ChatGLM, BiLLa(based on LLaMA)...These may be integrated sometime in the
future.
2023-08-18 20:49:50 +02:00