Merge branch 'main' into brace/cohere-token-count

docs/core_docs/.gitignore

@@ -176,4 +176,4 @@ docs/how_to/assign.mdx
 docs/how_to/agent_executor.md
 docs/how_to/agent_executor.mdx
 docs/integrations/llms/mistral.md
-docs/integrations/llms/mistral.mdx
+docs/integrations/llms/mistral.mdx

docs/core_docs/docs/integrations/chat/deep_infra.mdx

@@ -0,0 +1,25 @@
+---
+sidebar_label: Deep Infra
+---
+
+import CodeBlock from "@theme/CodeBlock";
+
+# ChatDeepInfra
+
+LangChain supports chat models hosted by [Deep Infra](https://deepinfra.com/) through the `ChatDeepInfra` wrapper.
+First, you'll need to install the `@langchain/community` package:
+
+import IntegrationInstallTooltip from "@mdx_components/integration_install_tooltip.mdx";
+
+<IntegrationInstallTooltip></IntegrationInstallTooltip>
+
+```bash npm2yarn
+npm install @langchain/community
+```
+
+You'll need to obtain an API key and set it as an environment variable named `DEEPINFRA_API_TOKEN`
+(or pass it into the constructor), then call the model as shown below:
+
+import Example from "@examples/models/chat/integration_deepinfra.ts";
+
+<CodeBlock language="typescript">{Example}</CodeBlock>

docs/core_docs/docs/integrations/chat/openai.mdx

@@ -138,3 +138,16 @@ import OpenAIStreamTokens from "@examples/models/chat/integration_openai_stream_
 :::tip
 See the LangSmith trace [here](https://smith.langchain.com/public/66bf7377-cc69-4676-91b6-25929a05e8b7/r)
 :::
+
+### Disabling parallel tool calls
+
+If you have multiple tools bound to the model, but you'd only like for a single tool to be called at a time, you can pass the `parallel_tool_calls` call option to enable/disable this behavior.
+By default, `parallel_tool_calls` is set to `true`.
+
+import OpenAIParallelToolCallsTokens from "@examples/models/chat/integration_openai_parallel_tool_calls.ts";
+
+<CodeBlock language="typescript">{OpenAIParallelToolCallsTokens}</CodeBlock>
+
+:::tip
+See the LangSmith trace for the first invocation [here](https://smith.langchain.com/public/68f2ff13-6331-47d8-a8c0-d1745788e84e/r) and the second invocation [here](https://smith.langchain.com/public/6c2fff29-9470-486a-8715-805fda631024/r)
+:::

docs/core_docs/docs/integrations/llms/deep_infra.mdx

@@ -0,0 +1,25 @@
+---
+sidebar_label: Deep Infra
+---
+
+import CodeBlock from "@theme/CodeBlock";
+
+# DeepInfra
+
+LangChain supports LLMs hosted by [Deep Infra](https://deepinfra.com/) through the `DeepInfra` wrapper.
+First, you'll need to install the `@langchain/community` package:
+
+import IntegrationInstallTooltip from "@mdx_components/integration_install_tooltip.mdx";
+
+<IntegrationInstallTooltip></IntegrationInstallTooltip>
+
+```bash npm2yarn
+npm install @langchain/community
+```
+
+You'll need to obtain an API key and set it as an environment variable named `DEEPINFRA_API_TOKEN`
+(or pass it into the constructor), then call the model as shown below:
+
+import Example from "@examples/models/llm/deepinfra.ts";
+
+<CodeBlock language="typescript">{Example}</CodeBlock>

examples/src/models/chat/integration_deepinfra.ts

@@ -0,0 +1,17 @@
+import { ChatDeepInfra } from "@langchain/community/chat_models/deepinfra";
+import { HumanMessage } from "@langchain/core/messages";
+
+const apiKey = process.env.DEEPINFRA_API_TOKEN;
+
+const model = "meta-llama/Meta-Llama-3-70B-Instruct";
+
+const chat = new ChatDeepInfra({
+  model,
+  apiKey,
+});
+
+const messages = [new HumanMessage("Hello")];
+
+const res = await chat.invoke(messages);
+
+console.log(res);

examples/src/models/chat/integration_openai_parallel_tool_calls.ts

@@ -0,0 +1,85 @@
+import { ChatOpenAI } from "@langchain/openai";
+import { z } from "zod";
+import { zodToJsonSchema } from "zod-to-json-schema";
+
+const model = new ChatOpenAI({
+  temperature: 0,
+  model: "gpt-4o",
+});
+
+// Define your tools
+const calculatorSchema = z
+  .object({
+    operation: z.enum(["add", "subtract", "multiply", "divide"]),
+    number1: z.number(),
+    number2: z.number(),
+  })
+  .describe("A tool to perform basic arithmetic operations");
+const weatherSchema = z
+  .object({
+    city: z.enum(["add", "subtract", "multiply", "divide"]),
+  })
+  .describe("A tool to get the weather in a city");
+
+// Bind tools to the model
+const modelWithTools = model.bindTools([
+  {
+    type: "function",
+    function: {
+      name: "calculator",
+      description: calculatorSchema.description,
+      parameters: zodToJsonSchema(calculatorSchema),
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "weather",
+      description: weatherSchema.description,
+      parameters: zodToJsonSchema(weatherSchema),
+    },
+  },
+]);
+
+// Invoke the model with `parallel_tool_calls` set to `true`
+const response = await modelWithTools.invoke(
+  ["What is the weather in san francisco and what is 23716 times 27342?"],
+  {
+    parallel_tool_calls: true,
+  }
+);
+console.log(response.tool_calls);
+// We can see it called two tools
+/*
+[
+  {
+    name: 'weather',
+    args: { city: 'san francisco' },
+    id: 'call_c1KymEIix7mdlFtgLSnTXmDc'
+  },
+  {
+    name: 'calculator',
+    args: { operation: 'multiply', number1: 23716, number2: 27342 },
+    id: 'call_ANLYclAmXQ4TwUCLXakbPr3Z'
+  }
+]
+*/
+
+// Invoke the model with `parallel_tool_calls` set to `false`
+const response2 = await modelWithTools.invoke(
+  ["What is the weather in san francisco and what is 23716 times 27342?"],
+  {
+    parallel_tool_calls: false,
+  }
+);
+console.log(response2.tool_calls);
+// We can see it called one tool
+/*
+[
+  {
+    name: 'weather',
+    args: { city: 'san francisco' },
+    id: 'call_Rk34XffawJjgZ2BCK9E4CwlT'
+  }
+]
+*/

examples/src/models/llm/deepinfra.ts

@@ -0,0 +1,18 @@
+import { DeepInfraLLM } from "@langchain/community/llms/deepinfra";
+
+const apiKey = process.env.DEEPINFRA_API_TOKEN;
+const model = "meta-llama/Meta-Llama-3-70B-Instruct";
+
+const llm = new DeepInfraLLM({
+  temperature: 0.7,
+  maxTokens: 20,
+  model,
+  apiKey,
+  maxRetries: 5,
+});
+
+const res = await llm.invoke(
+  "What is the next step in the process of making a good game?"
+);
+
+console.log({ res });

libs/langchain-community/.gitignore

@@ -234,6 +234,10 @@ llms/cohere.cjs
 llms/cohere.js
 llms/cohere.d.ts
 llms/cohere.d.cts
+llms/deepinfra.cjs
+llms/deepinfra.js
+llms/deepinfra.d.ts
+llms/deepinfra.d.cts
 llms/fireworks.cjs
 llms/fireworks.js
 llms/fireworks.d.ts
@@ -510,6 +514,10 @@ chat_models/cloudflare_workersai.cjs
 chat_models/cloudflare_workersai.js
 chat_models/cloudflare_workersai.d.ts
 chat_models/cloudflare_workersai.d.cts
+chat_models/deepinfra.cjs
+chat_models/deepinfra.js
+chat_models/deepinfra.d.ts
+chat_models/deepinfra.d.cts
 chat_models/fireworks.cjs
 chat_models/fireworks.js
 chat_models/fireworks.d.ts

libs/langchain-community/langchain.config.js

@@ -93,6 +93,7 @@ export const config = {
     "llms/bedrock/web": "llms/bedrock/web",
     "llms/cloudflare_workersai": "llms/cloudflare_workersai",
     "llms/cohere": "llms/cohere",
+    "llms/deepinfra": "llms/deepinfra",
     "llms/fireworks": "llms/fireworks",
     "llms/friendli": "llms/friendli",
     "llms/googlepalm": "llms/googlepalm",
@@ -164,6 +165,7 @@ export const config = {
     "chat_models/bedrock": "chat_models/bedrock/index",
     "chat_models/bedrock/web": "chat_models/bedrock/web",
     "chat_models/cloudflare_workersai": "chat_models/cloudflare_workersai",
+    "chat_models/deepinfra": "chat_models/deepinfra",
     "chat_models/fireworks": "chat_models/fireworks",
     "chat_models/friendli": "chat_models/friendli",
     "chat_models/googlevertexai": "chat_models/googlevertexai/index",

libs/langchain-community/package.json

@@ -1231,6 +1231,15 @@
       "import": "./llms/cohere.js",
       "require": "./llms/cohere.cjs"
     },
+    "./llms/deepinfra": {
+      "types": {
+        "import": "./llms/deepinfra.d.ts",
+        "require": "./llms/deepinfra.d.cts",
+        "default": "./llms/deepinfra.d.ts"
+      },
+      "import": "./llms/deepinfra.js",
+      "require": "./llms/deepinfra.cjs"
+    },
     "./llms/fireworks": {
       "types": {
         "import": "./llms/fireworks.d.ts",
@@ -1852,6 +1861,15 @@
       "import": "./chat_models/cloudflare_workersai.js",
       "require": "./chat_models/cloudflare_workersai.cjs"
     },
+    "./chat_models/deepinfra": {
+      "types": {
+        "import": "./chat_models/deepinfra.d.ts",
+        "require": "./chat_models/deepinfra.d.cts",
+        "default": "./chat_models/deepinfra.d.ts"
+      },
+      "import": "./chat_models/deepinfra.js",
+      "require": "./chat_models/deepinfra.cjs"
+    },
     "./chat_models/fireworks": {
       "types": {
         "import": "./chat_models/fireworks.d.ts",
@@ -3235,6 +3253,10 @@
     "llms/cohere.js",
     "llms/cohere.d.ts",
     "llms/cohere.d.cts",
+    "llms/deepinfra.cjs",
+    "llms/deepinfra.js",
+    "llms/deepinfra.d.ts",
+    "llms/deepinfra.d.cts",
     "llms/fireworks.cjs",
     "llms/fireworks.js",
     "llms/fireworks.d.ts",
@@ -3511,6 +3533,10 @@
     "chat_models/cloudflare_workersai.js",
     "chat_models/cloudflare_workersai.d.ts",
     "chat_models/cloudflare_workersai.d.cts",
+    "chat_models/deepinfra.cjs",
+    "chat_models/deepinfra.js",
+    "chat_models/deepinfra.d.ts",
+    "chat_models/deepinfra.d.cts",
     "chat_models/fireworks.cjs",
     "chat_models/fireworks.js",
     "chat_models/fireworks.d.ts",