@ax-llm/ax 9.0.0 → 9.0.2

package/README.md

@@ -1,28 +1,28 @@
 # Ax - Build LLMs Powered Agents (Typescript)
 
-JS/TS library to make to easy to build with agents and agentic workflows with LLMs. Full support for various LLMs and VectorDBs, Function Calling, Chain-of-Thought, RAG, Semantic Router and more. Based on the popular Stanford DSP paper. Build agents or teams of agents to solve complex problems.
+Build intelligent agents with ease, inspired by the power of "Agentic workflows" and the Stanford DSP paper. Seamlessly integrates with multiple LLMs and VectorDBs to build RAG pipelines or collaborative agents that can solve complex problems. Advanced features streaming validation, multi-modal DSP, etc.
 
 [![NPM Package](https://img.shields.io/npm/v/ax?style=for-the-badge&color=green)](https://www.npmjs.com/package/@ax-llm/ax)
 [![Twitter](https://img.shields.io/twitter/follow/dosco?style=for-the-badge&color=red)](https://twitter.com/dosco)
 [![Discord Chat](https://dcbadge.vercel.app/api/server/DSHg3dU7dW?style=for-the-badge)](https://discord.gg/DSHg3dU7dW)
 
-![llama-small](https://github.com/dosco/llm-client/assets/832235/b959fdd6-c723-49b1-9fb9-bf879e75c147)
+![image](https://github.com/ax-llm/ax/assets/832235/3a250031-591c-42e0-b4fc-06afb8c351c4)
 
-## Build with prompt signatures
+## Our focus on agents
 
-Ax is an easy to use library built around "Prompt Signatures" from the `Stanford DSP` paper. This library will automatically generate efficient and typed prompts from prompt signatures like `question:string -> answer:string`.
+We've renamed from "llmclient" to "ax" to highlight our focus on powering agentic workflows. We agree with many experts like "Andrew Ng" that agentic workflows are the key to unlocking the true power of large language models and what can be achieved with in-context learning. Also we are big fans of the Stanford DSP paper and this library is the result of all of this coming together to build a powerful framework for you to build with.
 
-Build powerful workflows using components like RAG, ReAcT, Chain of Thought, Function calling, Agents, etc all built on prompt signatures and easy to compose together to build whatever you want. Using prompt signatures automatically gives you the ability to fine tune your prompt programs using optimizers. Tune with a larger model and have your program run efficiently on a smaller model. The tuning here is not the traditional model tuning but what we call prompt tuning.
+![image](https://github.com/ax-llm/ax/assets/832235/801b8110-4cba-4c50-8ec7-4d5859121fe5)
 
 ## Why use Ax?
 
 - Support for various LLMs and Vector DBs
 - Prompts auto-generated from simple signatures
-- Multi-Hop RAG, ReAcT, CoT, Function Calling and more
 - Build Agents that can call other agents
 - Convert docs of any format to text
 - RAG, smart chunking, embedding, querying
-- Output field processing, validation while streaming
+- Output validation while streaming
+- Multi-modal DSP supported
 - Automatic prompt tuning using optimizers
 - OpenTelemetry tracing / observability
 - Production ready Typescript code
@@ -32,7 +32,7 @@ Build powerful workflows using components like RAG, ReAcT, Chain of Thought, Fun
 
 <img width="860" alt="shapes at 24-03-31 00 05 55" src="https://github.com/dosco/llm-client/assets/832235/0f0306ea-1812-4a0a-9ed5-76cd908cd26b">
 
-Efficient type-safe prompts are auto-generated from a simple signature. A prompt signature is made of a `"task description" inputField:type "field description" -> outputField:type"`. The idea behind prompt signatures is based off work done in the "Demonstrate-Search-Predict" paper.
+Efficient type-safe prompts are auto-generated from a simple signature. A prompt signature is made up of a `"task description" inputField:type "field description" -> outputField:type"`. The idea behind prompt signatures is based off work done in the "Demonstrate-Search-Predict" paper.
 
 You can have multiple input and output fields and each field has one of these types `string`, `number`, `boolean`, `json` or a array of any of these eg. `string[]`. When a type is not defined it defaults to `string`. When the `json` type if used the underlying AI is encouraged to generate correct JSON.
 
@@ -63,13 +63,13 @@ yarn add @ax-llm/ax
 ## Example: Using chain-of-thought to summarize text
 
 ```typescript
-import { AI, ChainOfThought, OpenAIArgs } from '@ax-llm/ax';
+import { axAI, AxChainOfThought } from '@ax-llm/ax';
 
 const textToSummarize = `
 The technological singularity—or simply the singularity[1]—is a hypothetical future point in time at which technological growth becomes uncontrollable and irreversible, resulting in unforeseeable changes to human civilization.[2][3] ...`;
 
-const ai = AI('openai', { apiKey: process.env.OPENAI_APIKEY } as OpenAIArgs);
-const gen = new ChainOfThought(
+const ai = axAI('openai', { apiKey: process.env.OPENAI_APIKEY });
+const gen = new AxChainOfThought(
   ai,
   `textToSummarize -> shortSummary "summarize in 5 to 10 words"`
 );
@@ -85,19 +85,19 @@ Use the agent prompt (framework) to build agents that work with other agents to
 ```typescript
 # npm run tsx ./src/examples/agent.ts
 
-const researcher = new Agent(ai, {
+const researcher = new AxAgent(ai, {
   name: 'researcher',
   description: 'Researcher agent',
   signature: `physicsQuestion "physics questions" -> answer "reply in bullet points"`
 });
 
-const summarizer = new Agent(ai, {
+const summarizer = new AxAgent(ai, {
   name: 'summarizer',
   description: 'Summarizer agent',
   signature: `text "text so summarize" -> shortSummary "summarize in 5 to 10 words"`
 });
 
-const agent = new Agent(ai, {
+const agent = new AxAgent(ai, {
   name: 'agent',
   description: 'A an agent to research complex topics',
   signature: `question -> answer`,
@@ -107,49 +107,6 @@ const agent = new Agent(ai, {
 agent.forward({ questions: "How many atoms are there in the universe" })
 ```
 
-## Fast LLM Router
-
-A special router that uses no LLM calls only embeddings to route user requests smartly.
-
-Use the Router to efficiently route user queries to specific routes designed to handle certain types of questions or tasks. Each route is tailored to a particular domain or service area. Instead of using a slow or expensive LLM to decide how input from the user should be handled use our fast "Semantic Router" that uses inexpensive and fast embedding queries.
-
-```typescript
-# npm run tsx ./src/examples/routing.ts
-
-const customerSupport = new Route('customerSupport', [
-  'how can I return a product?',
-  'where is my order?',
-  'can you help me with a refund?',
-  'I need to update my shipping address',
-  'my product arrived damaged, what should I do?'
-]);
-
-const technicalSupport = new Route('technicalSupport', [
-  'how do I install your software?',
-  'I’m having trouble logging in',
-  'can you help me configure my settings?',
-  'my application keeps crashing',
-  'how do I update to the latest version?'
-]);
-
-const ai = AI('openai', { apiKey: process.env.OPENAI_APIKEY } as OpenAIArgs);
-
-const router = new Router(ai);
-await router.setRoutes(
-  [customerSupport, technicalSupport],
-  { filename: 'router.json' }
-);
-
-const tag = await router.forward('I need help with my order');
-
-if (tag === "customerSupport") {
-    ...
-}
-if (tag === "technicalSupport") {
-    ...
-}
-```
-
 ## Vector DBs Supported
 
 Vector databases are critical to building LLM workflows. We have clean abstractions over popular vector db's as well as our own quick in memory vector database.
@@ -166,7 +123,7 @@ Vector databases are critical to building LLM workflows. We have clean abstracti
 const ret = await this.ai.embed({ texts: 'hello world' });
 
 // Create an in memory vector db
-const db = new DB('memory');
+const db = new axDB('memory');
 
 // Insert into vector db
 await this.db.upsert({
@@ -182,11 +139,11 @@ const matches = await this.db.query({
 });
 ```
 
-Alternatively you can use the `DBManager` which handles smart chunking, embedding and querying everything
+Alternatively you can use the `AxDBManager` which handles smart chunking, embedding and querying everything
 for you, it makes things almost too easy.
 
 ```typescript
-const manager = new DBManager({ ai, db });
+const manager = new AxDBManager({ ai, db });
 await manager.insert(text);
 
 const matches = await manager.query(
@@ -205,26 +162,43 @@ Launch Apache Tika
 docker run -p 9998:9998 apache/tika
 ```
 
-Convert documents to text and embed them for retrieval using the `DBManager` it also supports a reranker and query rewriter. Two default implementations `DefaultResultReranker` and `DefaultQueryRewriter` are available to use.
+Convert documents to text and embed them for retrieval using the `AxDBManager` it also supports a reranker and query rewriter. Two default implementations `AxDefaultResultReranker` and `AxDefaultQueryRewriter` are available to use.
 
 ```typescript
-const tika = new ApacheTika();
+const tika = new AxApacheTika();
 const text = await tika.convert('/path/to/document.pdf');
 
-const manager = new DBManager({ ai, db });
+const manager = new AxDBManager({ ai, db });
 await manager.insert(text);
 
 const matches = await manager.query('Find some text');
 console.log(matches);
 ```
 
+## Multi-modal DSP
+
+When using models like `gpt-4o` and `gemini` that support multi-modal prompts we support using image fields and this works with the whole dsp pipeline.
+
+```typescript
+const image = fs
+  .readFileSync('./src/examples/assets/kitten.jpeg')
+  .toString('base64');
+
+const gen = new AxChainOfThought(ai, `question, animalImage:image -> answer`);
+
+const res = await gen.forward({
+  question: 'What family does this animal belong to?',
+  animalImage: { mimeType: 'image/jpeg', data: image }
+});
+```
+
 ## Streaming
 
 We support parsing output fields and function execution while streaming. This allows for fail-fast and error correction without having to wait for the whole output saving tokens, cost and reducing latency. Assertions are a powerful way to ensure the output matches your requirements these work with streaming as well.
 
 ```typescript
 // setup the prompt program
-const gen = new ChainOfThought(
+const gen = new AxChainOfThought(
   ai,
   `startNumber:number -> next10Numbers:number[]`
 );
@@ -267,6 +241,49 @@ const res = await gen.forward(
 );
 ```
 
+## Fast LLM Router
+
+A special router that uses no LLM calls only embeddings to route user requests smartly.
+
+Use the Router to efficiently route user queries to specific routes designed to handle certain types of questions or tasks. Each route is tailored to a particular domain or service area. Instead of using a slow or expensive LLM to decide how input from the user should be handled use our fast "Semantic Router" that uses inexpensive and fast embedding queries.
+
+```typescript
+# npm run tsx ./src/examples/routing.ts
+
+const customerSupport = new AxRoute('customerSupport', [
+  'how can I return a product?',
+  'where is my order?',
+  'can you help me with a refund?',
+  'I need to update my shipping address',
+  'my product arrived damaged, what should I do?'
+]);
+
+const technicalSupport = new AxRoute('technicalSupport', [
+  'how do I install your software?',
+  'I’m having trouble logging in',
+  'can you help me configure my settings?',
+  'my application keeps crashing',
+  'how do I update to the latest version?'
+]);
+
+const ai = axAI('openai', { apiKey: process.env.OPENAI_APIKEY });
+
+const router = new AxRouter(ai);
+await router.setRoutes(
+  [customerSupport, technicalSupport],
+  { filename: 'router.json' }
+);
+
+const tag = await router.forward('I need help with my order');
+
+if (tag === "customerSupport") {
+    ...
+}
+if (tag === "technicalSupport") {
+    ...
+}
+```
+
 ## OpenTelemetry support
 
 Ability to trace and observe your llm workflow is critical to building production workflows. OpenTelemetry is an industry standard and we support the new `gen_ai` attribute namespace.
@@ -285,12 +302,12 @@ trace.setGlobalTracerProvider(provider);
 
 const tracer = trace.getTracer('test');
 
-const ai = AI('ollama', {
+const ai = axAI('ollama', {
   model: 'nous-hermes2',
   options: { tracer }
 } as unknown as OllamaArgs);
 
-const gen = new ChainOfThought(
+const gen = new AxChainOfThought(
   ai,
   `text -> shortSummary "summarize in 5 to 10 words"`
 );
@@ -322,26 +339,13 @@ const res = await gen.forward({ text });
 }
 ```
 
-Alternatively you can use the `DBManager` which handles smart chunking, embedding and querying everything
-for you, it makes things almost too easy.
-
-```typescript
-const manager = new DBManager({ ai, db });
-await manager.insert(text);
-
-const matches = await manager.query(
-  'John von Neumann on human intelligence and singularity.'
-);
-console.log(matches);
-```
-
 ## Tuning the prompts (programs)
 
-You can tune your prompts using a larger model to help them run more efficiently and give you better results. This is done by using an optimizer like `BootstrapFewShot` with and examples from the popular `HotPotQA` dataset. The optimizer generates demonstrations `demos` which when used with the prompt help improve its efficiency.
+You can tune your prompts using a larger model to help them run more efficiently and give you better results. This is done by using an optimizer like `AxBootstrapFewShot` with and examples from the popular `HotPotQA` dataset. The optimizer generates demonstrations `demos` which when used with the prompt help improve its efficiency.
 
 ```typescript
 // Download the HotPotQA dataset from huggingface
-const hf = new HFDataLoader();
+const hf = new AxHFDataLoader();
 const examples = await hf.getData<{ question: string; answer: string }>({
   dataset: 'hotpot_qa',
   split: 'train',
@@ -349,24 +353,25 @@ const examples = await hf.getData<{ question: string; answer: string }>({
   fields: ['question', 'answer']
 });
 
-const ai = AI('openai', { apiKey: process.env.OPENAI_APIKEY } as OpenAIArgs);
+const ai = axAI('openai', { apiKey: process.env.OPENAI_APIKEY });
 
 // Setup the program to tune
-const program = new ChainOfThought<{ question: string }, { answer: string }>(
+const program = new AxChainOfThought<{ question: string }, { answer: string }>(
   ai,
   `question -> answer "in short 2 or 3 words"`
 );
 
 // Setup a Bootstrap Few Shot optimizer to tune the above program
-const optimize = new BootstrapFewShot<{ question: string }, { answer: string }>(
-  {
-    program,
-    examples
-  }
-);
+const optimize = new AxBootstrapFewShot<
+  { question: string },
+  { answer: string }
+>({
+  program,
+  examples
+});
 
 // Setup a evaluation metric em, f1 scores are a popular way measure retrieval performance.
-const metricFn: MetricFn = ({ prediction, example }) =>
+const metricFn: AxMetricFn = ({ prediction, example }) =>
   emScore(prediction.answer as string, example.answer as string);
 
 // Run the optimizer and save the result
@@ -378,10 +383,10 @@ await optimize.compile(metricFn, { filename: 'demos.json' });
 And to use the generated demos with the above `ChainOfThought` program
 
 ```typescript
-const ai = AI('openai', { apiKey: process.env.OPENAI_APIKEY } as OpenAIArgs);
+const ai = axAI('openai', { apiKey: process.env.OPENAI_APIKEY });
 
 // Setup the program to use the tuned data
-const program = new ChainOfThought<{ question: string }, { answer: string }>(
+const program = new AxChainOfThought<{ question: string }, { answer: string }>(
   ai,
   `question -> answer "in short 2 or 3 words"`
 );
@@ -421,6 +426,7 @@ OPENAI_APIKEY=openai_key npm run tsx ./src/examples/marketing.ts
 | qna-use-tuned.ts    | Use the optimized tuned prompts                         |
 | streaming1.ts       | Output fields validation while streaming                |
 | streaming2.ts       | Per output field validation while streaming             |
+| smart-hone.ts       | Agent looks for dog in smart home                       |
 
 ## Built-in Functions
 
@@ -439,7 +445,7 @@ Large language models (LLMs) are getting really powerful and have reached a poin
 
 ```ts
 // Pick a LLM
-const ai = new OpenAI({ apiKey: process.env.OPENAI_APIKEY } as OpenAIArgs);
+const ai = new AxOpenAI({ apiKey: process.env.OPENAI_APIKEY } as AxOpenAIArgs);
 ```
 
 ### 2. Create a prompt signature based on your usecase
@@ -501,13 +507,13 @@ const functions = [
 ### 2. Pass the functions to a prompt
 
 ```ts
-const cot = new ReAct(ai, `question:string -> answer:string`, { functions });
+const cot = new AxReAct(ai, `question:string -> answer:string`, { functions });
 ```
 
 ## Enable debug logs
 
 ```ts
-const ai = new OpenAI({ apiKey: process.env.OPENAI_APIKEY } as OpenAIArgs);
+const ai = new AxOpenAI({ apiKey: process.env.OPENAI_APIKEY } as AxOpenAIArgs);
 ai.setOptions({ debug: true });
 ```
 
@@ -528,20 +534,20 @@ You can pass a configuration object as the second parameter when creating a new
 
 ```ts
 const apiKey = process.env.OPENAI_APIKEY;
-const conf = OpenAIBestConfig();
-const ai = new OpenAI({ apiKey, conf } as OpenAIArgs);
+const conf = AxOpenAIBestConfig();
+const ai = new AxOpenAI({ apiKey, conf } as AxOpenAIArgs);
 ```
 
 ## 3. My prompt is too long and can I change the max tokens
 
 ```ts
-const conf = OpenAIDefaultConfig(); // or OpenAIBestOptions()
+const conf = axOpenAIDefaultConfig(); // or OpenAIBestOptions()
 conf.maxTokens = 2000;
 ```
 
 ## 4. How do I change the model say I want to use GPT4
 
 ```ts
-const conf = OpenAIDefaultConfig(); // or OpenAIBestOptions()
+const conf = axOpenAIDefaultConfig(); // or OpenAIBestOptions()
 conf.model = OpenAIModel.GPT4Turbo;
 ```

package/build/module/docs/assets/navigation.js

@@ -1,3 +1,3 @@
-window.navigationData = "data:application/octet-stream;base64,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";
+window.navigationData = "data:application/octet-stream;base64,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";
 export {};
 //# sourceMappingURL=navigation.js.map

package/build/module/docs/assets/navigation.js.map

@@ -1 +1 @@
-{"version":3,"file":"navigation.js","sourceRoot":"","sources":["../../../../docs/assets/navigation.js"],"names":[],"mappings":"AAAA,MAAM,CAAC,cAAc,GAAG,m4GAAm4G,CAAA"}
+{"version":3,"file":"navigation.js","sourceRoot":"","sources":["../../../../docs/assets/navigation.js"],"names":[],"mappings":"AAAA,MAAM,CAAC,cAAc,GAAG,+kGAA+kG,CAAA"}