FastGPT is a knowledge-based platform built on the LLMs, offers a comprehensive suite of out-of-the-box capabilities such as data processing, RAG retrieval, and visual AI workflow orchestration, letting you easily develop and deploy complex question-answering systems without the need for extensive setup or configuration.
4.9.10-fix
使用ollama qwen2:0.5b模型,fastgpt中在AI对话中调试就ok,oneapi日志如下 [INFO] 2024/09/14 - 15:25:20 | 2024091415252077050614860574139 | user 1 has enough quota 499999218863848, trusted and no need to pre-consume [GIN] 2024/09/14 - 15:25:21 | 2024091415252077050614860574139 | 200 | 479.207677ms | 172.19.0.1 | POST /v1/chat/completions [INFO] 2024/09/14 - 15:25:21 | 2024091415252077050614860574139 | record consume log: userId=1, channelId=10, promptTokens=0, completionTokens=18, modelName=qwen2:0.5b, tokenName=OllamaJPLingPai, quota=540, content=模型倍率 30.00,分组倍率 1.00,补全倍率 1.00
但是直接运行应用就报错,oneapi报错如下
[INFO] 2024/09/14 - 15:22:09 | 2024091415220978606880125741390 | user 1 has enough quota 499999218865228, trusted and no need to pre-consume [SYS] 2024/09/14 - 15:22:09 | error unmarshalling stream response: invalid character '}' after top-level value [SYS] 2024/09/14 - 15:22:09 | error unmarshalling stream response: invalid character '}' after top-level value [SYS] 2024/09/14 - 15:22:09 | error unmarshalling stream response: invalid character '}' after top-level value [SYS] 2024/09/14 - 15:22:10 | error unmarshalling stream response: invalid character '}' after top-level value [SYS] 2024/09/14 - 15:22:10 | error unmarshalling stream response: invalid character '}' after top-level value [SYS] 2024/09/14 - 15:22:10 | error unmarshalling stream response: invalid character '}' after top-level value [SYS] 2024/09/14 - 15:22:10 | error unmarshalling stream response: invalid character '}' after top-level value [GIN] 2024/09/14 - 15:22:10 | 2024091415220978606880125741390 | 200 | 275.067781ms | 172.19.0.1 | POST /v1/chat/completions [INFO] 2024/09/14 - 15:22:10 | 2024091415220978606880125741390 | record consume log: userId=1, channelId=10, promptTokens=0, completionTokens=0, modelName=qwen2:0.5b, tokenName=OllamaJPLingPai, quota=0, content=模型倍率 30.00,分组倍率 1.00,补全倍率 1.00