Production-gradeLLM tools
programmatically
Due to the powerful understanding and generation capabilities of this wave of LLMs, command-style programming that pays attention to details seems to be no longer needed, and Statement programming that focuses on process or pipeline capabilities of business logic orchestration has become the mainstream "programming" method. Therefore, we have seen that rookies including LCEL, DSPy, and Semantic-kernel have begun to pay attention "upwards" and focus their main energy on business logic orchestration capabilities. Perhaps this indicates a DAG approach similar to Airflow. Will it become a new generation of programming normal form?
totality
Prompt tool
LangSmith or Azure Prompt Flow
Programming language
Python or Javascript
Development framework/orchestration tool
Langchain or Semitic Kernel
API framework/service box frame work
FastAPI or Next.js
Cloud hosting vendor
Render (AWS Lamada) or Vercel
BuildLLM applications
classification | Name | Star Quantity | remark | Learning |
---|---|---|---|---|
frame | LangChain | 77 | [Required] | * |
LlamaIndex | 29 | 【必学】专注于RAG,Python,JS SDK比较差 | * | |
Semantic-kernel | 17 | 【不确定?】C#,Python,Java | ? | |
haystack | 13 | 有了LangChain可能就不需要这个了 | ||
7 | Too academic? | 太学术了?The framework for programming-not prompting-foundation models | ||
Agent | SuperAgent | 4 | This feels very professional, and I will Learning again when I have time | ? |
Rift | 3 | 专门针对程序员 | ||
AutoGen | 22 | |||
Camel | 4 | |||
GPTs | OpenGPTs | 6 | Produced by LangChain | |
low-code | LangFlow | LangChain visualization | ||
LiteLLM | ||||
Flowise | 21 | |||
Langflow | 14 | |||
PromptChainer | 商业 | 感觉做的很专业 | ||
RAG | vectara | closed source | Very professional RAG commercial soft ware; there are also Open-Source projects; users can upload PDF, Word, PPT, RTF and other file Data to the Vectara platform to build a Data Searching engine. |
Archiving and Observation
classification | Name | Star Quantity | remark |
---|---|---|---|
observe | Obsidian Copilot | An interesting approach to how to use Semantic Searching and the BM25 implementation of OpenSearch | |
Tanuki | 一个使用装饰器进行数据验证的 LLM 框架 | ||
griptape | 1.5 | 一个具有稍微更好的内部编码标准的 langchain 替代品 | |
16 | Controlling modern Language Model more Valid Earth than traditional prompts or links | 比传统的提示或链接更有效地控制现代语言模型 | |
intructor | 3.6 | ||
0.5 | Another alternative to langchain, but has an interesting approach to defining workflows | langchain 的另一个替代品,但在定义工作流程方面有一个有趣的方法 | |
6.5 | There are derivative products for chat, medical/scientific paper workflow, developer semantic Searching, and title and story text semantic Searching | 有用于聊天、医学/科学论文工作流程、开发者语义搜索以及标题和故事文本语义搜索的衍生产品 | |
5.7 | A simple, lightweightLLM package | 一个简单、轻量级的 LLM 封装器 | |
pigeonhole | Chidori | 1.2 | |
AutoChains | 已经3个月没更新 | ||
MiniChain | 1.2 | 已经3个月没更新 | |
SimpleAIChat | 3.3 | 已经7个月没更新 |
monitor and analyze
LLM services at the production level require:
-
Debugging Prompt
-
Prompt version management
-
Metric related to the dependence of the test/verification system
-
Data Set Management
-
Various Metric monitoring and statistics: visit volume, response time, Token fee, etc.
classification | Name | Star Quantity | remark |
---|---|---|---|
SaaS | LangSmith | N/A | |
Open-Source | LangFuse | 2 | |
Open-Source | Prompt Flow | 7 | |
Helicone | 1.2 |
Production-levelLLM App maintenance platform
- LangSmith: LangChain's official platform, SaaS service,not Open-Source
- LangFuse:Open-Source+ SaaS, LangSmith replacement, Ensemble LangChain integration, docking with OpenAI API(2.1K)
- Prompt Flow: MicrosoftCorp Open-Source + Azure AI cloud service, which can Ensemble Semantic Kernel (7.4K)
Prompt Flow
Learning
https://microsoft.github.io/promptflow/how-to-guides/quick-start.html
UI component
LangUI(1.6K)
agentlabs(0.5K)
other
classification | Name | Star Quantity | remark |
---|---|---|---|
LCEL | |||
Python template language | Jinja2 | ||
reference
classification | Name | Star Quantity | remark | |
---|---|---|---|---|
frame | Obsidian Copilot | An interesting approach to how to use Semantic Searching and the BM25 implementation of OpenSearch | ||
Tanuki | 个使用装饰器进行数据验证的 LLM 框架 | |||
LiteLLM | 5.7 | 一个简单、轻量级的 LLM 封装器 | ||
AutoChain | 1.7 | langchain 的另一个替代品 | ||
griptape | 1.5 | 一个具有稍微更好的内部编码标准的 langchain 替代品 | ||
txtai | 6.5 | has spinoffs for chat, workflows for medical/scientific papers, semantic search for developers and semantic search for headlines and story text 有用于聊天、医学/科学论文工作流程、开发者语义搜索以及标题和故事文本语义搜索的衍生产品 | ||
Llmflows | 0.5 | langchain 的另一个替代品,但在定义工作流程方面有一个有趣的方法 | ||
guidance | 16 | 比传统的提示或链接更有效地控制现代语言模型 | * | |
library | intructor | 3.6 | * |
E2B:Frameworks and tools for AI products
Category | Tools/Platforms |
---|---|
Monitoring, Observability, Analytics | Langsmith, Helicone, langfuse, context |
Frontend | AgentLabs |
Runtime for LLMs | E2B |
Building frameworks & platforms | Langchain, Haystack by deepset, Superagent, Poe, GenWorlds, LiteLLM, Rift by Morph, AutoGen, Vercel AI SDK, OpenGPTs, Huggging Face Agents, Fixie |
Data integration, memory management | vectara, LlamaIndex, ABACUS.AI, cadea, SID Memory |
API and routers for LLMs | Genoss GPT, Martian, OpenRouter |
Libraries for building AI products | CAMEL, ChatterBot (hasn't been updated in 5 annum) |
Orchestration | AGIXT |
Building & deploying LLMs | BANANA |