github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #2513] ECONNREFUSED error #1468

New Issue
Closed
opened 2026-04-12 11:22:23 -05:00 by GiteaMirror · 17 comments
GiteaMirror commented 2026-04-12 11:22:23 -05:00
Owner
Copy Link

Originally created by @jakobhoeg on GitHub (Feb 15, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/2513

Keep getting ECONNREFUSED error when trying to use Ollama for my NextJS frontend in production:

⨯ TypeError: fetch failed
    at Object.fetch (node:internal/deps/undici/undici:11730:11)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
    at async globalThis.fetch (/var/task/node_modules/next/dist/compiled/next-server/app-route.runtime.prod.js:6:36091)
    at async s (/var/task/.next/server/app/api/model/route.js:1:491)
    at async /var/task/node_modules/next/dist/compiled/next-server/app-route.runtime.prod.js:6:42484
    at async eI.execute (/var/task/node_modules/next/dist/compiled/next-server/app-route.runtime.prod.js:6:32486)
    at async eI.handle (/var/task/node_modules/next/dist/compiled/next-server/app-route.runtime.prod.js:6:43737)
    at async Y (/var/task/node_modules/next/dist/compiled/next-server/server.runtime.prod.js:16:24556)
    at async Q.responseCache.get.routeKind (/var/task/node_modules/next/dist/compiled/next-server/server.runtime.prod.js:17:1025)
    at async r3.renderToResponseWithComponentsImpl (/var/task/node_modules/next/dist/compiled/next-server/server.runtime.prod.js:17:507) {
  cause: Error: connect ECONNREFUSED 127.0.0.1:11434
      at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1555:16)
      at TCPConnectWrap.callbackTrampoline (node:internal/async_hooks:128:17) {
    errno: -111,
    code: 'ECONNREFUSED',
    syscall: 'connect',
    address: '127.0.0.1',
    port: 11434
  }
}
Originally created by @jakobhoeg on GitHub (Feb 15, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/2513 Keep getting ECONNREFUSED error when trying to use Ollama for my NextJS frontend in production: ``` ⨯ TypeError: fetch failed at Object.fetch (node:internal/deps/undici/undici:11730:11) at process.processTicksAndRejections (node:internal/process/task_queues:95:5) at async globalThis.fetch (/var/task/node_modules/next/dist/compiled/next-server/app-route.runtime.prod.js:6:36091) at async s (/var/task/.next/server/app/api/model/route.js:1:491) at async /var/task/node_modules/next/dist/compiled/next-server/app-route.runtime.prod.js:6:42484 at async eI.execute (/var/task/node_modules/next/dist/compiled/next-server/app-route.runtime.prod.js:6:32486) at async eI.handle (/var/task/node_modules/next/dist/compiled/next-server/app-route.runtime.prod.js:6:43737) at async Y (/var/task/node_modules/next/dist/compiled/next-server/server.runtime.prod.js:16:24556) at async Q.responseCache.get.routeKind (/var/task/node_modules/next/dist/compiled/next-server/server.runtime.prod.js:17:1025) at async r3.renderToResponseWithComponentsImpl (/var/task/node_modules/next/dist/compiled/next-server/server.runtime.prod.js:17:507) { cause: Error: connect ECONNREFUSED 127.0.0.1:11434 at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1555:16) at TCPConnectWrap.callbackTrampoline (node:internal/async_hooks:128:17) { errno: -111, code: 'ECONNREFUSED', syscall: 'connect', address: '127.0.0.1', port: 11434 } } ```
GiteaMirror added the bugjsneeds more info labels 2026-04-12 11:22:24 -05:00
GiteaMirror commented 2026-04-12 11:22:25 -05:00
Author
Owner
Copy Link

@mxyng commented on GitHub (Feb 15, 2024):

ECONNREFUSED indicates Ollama server isn't running. Can you check it is running and accessible on localhost:11434?

<!-- gh-comment-id:1947137756 --> @mxyng commented on GitHub (Feb 15, 2024): ECONNREFUSED indicates Ollama server isn't running. Can you check it is running and accessible on localhost:11434?
GiteaMirror commented 2026-04-12 11:22:25 -05:00
Author
Owner
Copy Link

@jakobhoeg commented on GitHub (Feb 15, 2024):

ECONNREFUSED indicates Ollama server isn't running. Can you check it is running and accessible on localhost:11434?

It is running and accessible.

<!-- gh-comment-id:1947343222 --> @jakobhoeg commented on GitHub (Feb 15, 2024): > ECONNREFUSED indicates Ollama server isn't running. Can you check it is running and accessible on localhost:11434? It is running and accessible.
GiteaMirror commented 2026-04-12 11:22:26 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Mar 12, 2024):

@jakobhoeg Did you manage to get this working? @mxyng 's point is correct; maybe you were running it on a different port or a different host?

<!-- gh-comment-id:1992747589 --> @pdevine commented on GitHub (Mar 12, 2024): @jakobhoeg Did you manage to get this working? @mxyng 's point is correct; maybe you were running it on a different port or a different host?
GiteaMirror commented 2026-04-12 11:22:26 -05:00
Author
Owner
Copy Link

@jakobhoeg commented on GitHub (Mar 17, 2024):

@jakobhoeg Did you manage to get this working? @mxyng 's point is correct; maybe you were running it on a different port or a different host?

Hey.
No, still not working. It's hosted on Vercel and I have set the OLLAMA_ORIGINS.
Dev server it works fine, just not when trying to host it.

<!-- gh-comment-id:2002430693 --> @jakobhoeg commented on GitHub (Mar 17, 2024): > @jakobhoeg Did you manage to get this working? @mxyng 's point is correct; maybe you were running it on a different port or a different host? Hey. No, still not working. It's hosted on Vercel and I have set the OLLAMA_ORIGINS. Dev server it works fine, just not when trying to host it.
GiteaMirror commented 2026-04-12 11:22:27 -05:00
Author
Owner
Copy Link

@mxyng commented on GitHub (Mar 17, 2024):

Can you clarify where everything is deployed? You mentioned something is deployed in Vercel but the wording is vague. I assume the NextJS app you're calling Ollama from. If this is the case, 127.0.0.1 is probably not the right OLLAMA_HOST since that would be the Vercel deployment.

<!-- gh-comment-id:2002501436 --> @mxyng commented on GitHub (Mar 17, 2024): Can you clarify where everything is deployed? You mentioned something is deployed in Vercel but the wording is vague. I assume the NextJS app you're calling Ollama from. If this is the case, 127.0.0.1 is probably not the right OLLAMA_HOST since that would be the Vercel deployment.
GiteaMirror commented 2026-04-12 11:22:27 -05:00
Author
Owner
Copy Link

@jakobhoeg commented on GitHub (Mar 17, 2024):

Can you clarify where everything is deployed? You mentioned something is deployed in Vercel but the wording is vague. I assume the NextJS app you're calling Ollama from. If this is the case, 127.0.0.1 is probably not the right OLLAMA_HOST since that would be the Vercel deployment.

Yeah, my apologies.
I have my NextJS frontend deployed on Vercel. I am trying to allow users to chat with their own Ollama server running on their machine.
I am able to handle this client side, but not in an api route. That is where I get the error code from the original post. And now that I think of it, this probably isn't a Ollama issue, but rather an issue with the langchain package or Vercel/NextJS

For reference, this is my api route: (api/chattest)

import { StreamingTextResponse, Message } from "ai";
import { ChatOllama } from "@langchain/community/chat_models/ollama";
import { AIMessage, HumanMessage } from "@langchain/core/messages";
import { BytesOutputParser } from "@langchain/core/output_parsers";


export async function POST(req: Request) {
  const { messages, selectedModel } = await req.json();

  const model = new ChatOllama({
    baseUrl: "http://localhost:11434",
    model: selectedModel,
  });

  const parser = new BytesOutputParser();

  const stream = await model
    .pipe(parser)
    .stream(
      (messages as Message[]).map((m) =>
        m.role == "user"
          ? new HumanMessage(m.content)
          : new AIMessage(m.content)
      )
    );

    console.log(stream);

  return new StreamingTextResponse(stream);
}

Whenever I call that api endpoint, it also doesnt appear in the Ollama logs with any error message.

If i however try and do this client side, it works. Like this (just a quick example):

useEffect(() => {
    const fetchChat = async () => {
      const res = await fetch("http://localhost:11434/api/chat", {
        method: "POST",
        headers: {
          "Content-Type": "application/json",
        },
        body: JSON.stringify({ model: "gemma:2b", "messages": [
          {
            "role": "user",
            "content": "why is the sky blue?"
          }
        ] }),
      });
      const data = await res.json();
      console.log("Data from fetchChat:", data);

    }
    fetchChat();
  }
  , []);
<!-- gh-comment-id:2002533534 --> @jakobhoeg commented on GitHub (Mar 17, 2024): > Can you clarify where everything is deployed? You mentioned something is deployed in Vercel but the wording is vague. I assume the NextJS app you're calling Ollama from. If this is the case, 127.0.0.1 is probably not the right OLLAMA_HOST since that would be the Vercel deployment. Yeah, my apologies. I have my NextJS frontend deployed on Vercel. I am trying to allow users to chat with their own Ollama server running on their machine. I am able to handle this client side, but not in an api route. That is where I get the error code from the original post. And now that I think of it, this probably isn't a Ollama issue, but rather an issue with the langchain package or Vercel/NextJS For reference, this is my api route: (api/chattest) ```ts import { StreamingTextResponse, Message } from "ai"; import { ChatOllama } from "@langchain/community/chat_models/ollama"; import { AIMessage, HumanMessage } from "@langchain/core/messages"; import { BytesOutputParser } from "@langchain/core/output_parsers"; export async function POST(req: Request) { const { messages, selectedModel } = await req.json(); const model = new ChatOllama({ baseUrl: "http://localhost:11434", model: selectedModel, }); const parser = new BytesOutputParser(); const stream = await model .pipe(parser) .stream( (messages as Message[]).map((m) => m.role == "user" ? new HumanMessage(m.content) : new AIMessage(m.content) ) ); console.log(stream); return new StreamingTextResponse(stream); } ``` Whenever I call that api endpoint, it also doesnt appear in the Ollama logs with any error message. If i however try and do this client side, it works. Like this (just a quick example): ```tsx useEffect(() => { const fetchChat = async () => { const res = await fetch("http://localhost:11434/api/chat", { method: "POST", headers: { "Content-Type": "application/json", }, body: JSON.stringify({ model: "gemma:2b", "messages": [ { "role": "user", "content": "why is the sky blue?" } ] }), }); const data = await res.json(); console.log("Data from fetchChat:", data); } fetchChat(); } , []); ```
GiteaMirror commented 2026-04-12 11:22:28 -05:00
Author
Owner
Copy Link

@BruceMacD commented on GitHub (Mar 18, 2024):

@jakobhoeg looks like this could be an issue with host resolution when using the langchain library rather than ollama, could you try using 127.0.0.1 rather than localhost?

const model = new ChatOllama({
    baseUrl: "http://127.0.0.1:11434",
    model: selectedModel,
  });
<!-- gh-comment-id:2003277007 --> @BruceMacD commented on GitHub (Mar 18, 2024): @jakobhoeg looks like this could be an issue with host resolution when using the langchain library rather than ollama, could you try using `127.0.0.1` rather than `localhost`? ```javascript const model = new ChatOllama({ baseUrl: "http://127.0.0.1:11434", model: selectedModel, }); ```
GiteaMirror commented 2026-04-12 11:22:28 -05:00
Author
Owner
Copy Link

@jakobhoeg commented on GitHub (Mar 18, 2024):

@jakobhoeg looks like this could be an issue with host resolution when using the langchain library rather than ollama, could you try using 127.0.0.1 rather than localhost?

const model = new ChatOllama({
    baseUrl: "http://127.0.0.1:11434",
    model: selectedModel,
  });

Have tried that. The issue persists.

<!-- gh-comment-id:2003326497 --> @jakobhoeg commented on GitHub (Mar 18, 2024): > @jakobhoeg looks like this could be an issue with host resolution when using the langchain library rather than ollama, could you try using `127.0.0.1` rather than `localhost`? > > ```js > const model = new ChatOllama({ > baseUrl: "http://127.0.0.1:11434", > model: selectedModel, > }); > ``` Have tried that. The issue persists.
GiteaMirror commented 2026-04-12 11:22:28 -05:00
Author
Owner
Copy Link

@jakobhoeg commented on GitHub (Mar 18, 2024):

Also tried with this in the api route:

export const runtime = 'edge';

And now I get a different error in Vercel:

 Error: Ollama call failed with status code 403: Direct IP access is not allowed in Vercel's Edge environment (hostname: 127.0.0.1)
    at (node_modules/@langchain/community/dist/utils/ollama.js:26:20)
    at (node_modules/@langchain/community/dist/utils/ollama.js:56:4)
    at (node_modules/@langchain/community/dist/chat_models/ollama.js:375:29)
    at (node_modules/@langchain/community/dist/chat_models/ollama.js:483:25)
    at (node_modules/@langchain/core/dist/language_models/chat_models.js:347:21)
    at (node_modules/@langchain/core/dist/language_models/chat_models.js:110:24)
    at (node_modules/@langchain/core/dist/language_models/chat_models.js:294:23) {
  response: Response {  }
}

This is basically because of how @langchain/community/dist/utils/ollama.js formats localhost to 127.0.0.1:

import { IterableReadableStream } from "@langchain/core/utils/stream";
async function* createOllamaStream(url, params, options) {
    let formattedUrl = url;
    if (formattedUrl.startsWith("http://localhost:")) {
        // Node 18 has issues with resolving "localhost"
        // See https://github.com/node-fetch/node-fetch/issues/1624
        formattedUrl = formattedUrl.replace("http://localhost:", "http://127.0.0.1:");
    }
    const response = await fetch(formattedUrl, {
        method: "POST",
        body: JSON.stringify(params),
        headers: {
            "Content-Type": "application/json",
        },
        signal: options.signal,
    });
    if (!response.ok) {
        let error;
        const responseText = await response.text();
        try {
            const json = JSON.parse(responseText);
            error = new Error(`Ollama call failed with status code ${response.status}: ${json.error}`);
            // eslint-disable-next-line @typescript-eslint/no-explicit-any
        }
        catch (e) {
            error = new Error(`Ollama call failed with status code ${response.status}: ${responseText}`);
        }
        // eslint-disable-next-line @typescript-eslint/no-explicit-any
        error.response = response;
        throw error;
    }
    if (!response.body) {
        throw new Error("Could not begin Ollama stream. Please check the given URL and try again.");
    }
    const stream = IterableReadableStream.fromReadableStream(response.body);
    const decoder = new TextDecoder();
    let extra = "";
    for await (const chunk of stream) {
        const decoded = extra + decoder.decode(chunk);
        const lines = decoded.split("\n");
        extra = lines.pop() || "";
        for (const line of lines) {
            try {
                yield JSON.parse(line);
            }
            catch (e) {
                console.warn(`Received a non-JSON parseable chunk: ${line}`);
            }
        }
    }
}
export async function* createOllamaGenerateStream(baseUrl, params, options) {
    yield* createOllamaStream(`${baseUrl}/api/generate`, params, options);
}
export async function* createOllamaChatStream(baseUrl, params, options) {
    yield* createOllamaStream(`${baseUrl}/api/chat`, params, options);
}

But I don't know if it would work, if it didn't format it?

<!-- gh-comment-id:2003500010 --> @jakobhoeg commented on GitHub (Mar 18, 2024): Also tried with this in the api route: `export const runtime = 'edge';` And now I get a different error in Vercel: ``` Error: Ollama call failed with status code 403: Direct IP access is not allowed in Vercel's Edge environment (hostname: 127.0.0.1) at (node_modules/@langchain/community/dist/utils/ollama.js:26:20) at (node_modules/@langchain/community/dist/utils/ollama.js:56:4) at (node_modules/@langchain/community/dist/chat_models/ollama.js:375:29) at (node_modules/@langchain/community/dist/chat_models/ollama.js:483:25) at (node_modules/@langchain/core/dist/language_models/chat_models.js:347:21) at (node_modules/@langchain/core/dist/language_models/chat_models.js:110:24) at (node_modules/@langchain/core/dist/language_models/chat_models.js:294:23) { response: Response { } } ``` This is basically because of how `@langchain/community/dist/utils/ollama.js` formats localhost to 127.0.0.1: ``` import { IterableReadableStream } from "@langchain/core/utils/stream"; async function* createOllamaStream(url, params, options) { let formattedUrl = url; if (formattedUrl.startsWith("http://localhost:")) { // Node 18 has issues with resolving "localhost" // See https://github.com/node-fetch/node-fetch/issues/1624 formattedUrl = formattedUrl.replace("http://localhost:", "http://127.0.0.1:"); } const response = await fetch(formattedUrl, { method: "POST", body: JSON.stringify(params), headers: { "Content-Type": "application/json", }, signal: options.signal, }); if (!response.ok) { let error; const responseText = await response.text(); try { const json = JSON.parse(responseText); error = new Error(`Ollama call failed with status code ${response.status}: ${json.error}`); // eslint-disable-next-line @typescript-eslint/no-explicit-any } catch (e) { error = new Error(`Ollama call failed with status code ${response.status}: ${responseText}`); } // eslint-disable-next-line @typescript-eslint/no-explicit-any error.response = response; throw error; } if (!response.body) { throw new Error("Could not begin Ollama stream. Please check the given URL and try again."); } const stream = IterableReadableStream.fromReadableStream(response.body); const decoder = new TextDecoder(); let extra = ""; for await (const chunk of stream) { const decoded = extra + decoder.decode(chunk); const lines = decoded.split("\n"); extra = lines.pop() || ""; for (const line of lines) { try { yield JSON.parse(line); } catch (e) { console.warn(`Received a non-JSON parseable chunk: ${line}`); } } } } export async function* createOllamaGenerateStream(baseUrl, params, options) { yield* createOllamaStream(`${baseUrl}/api/generate`, params, options); } export async function* createOllamaChatStream(baseUrl, params, options) { yield* createOllamaStream(`${baseUrl}/api/chat`, params, options); } ``` But I don't know if it would work, if it didn't format it?
GiteaMirror commented 2026-04-12 11:22:29 -05:00
Author
Owner
Copy Link

@BruceMacD commented on GitHub (Mar 18, 2024):

@jakobhoeg could this code be executing on the server-side rather than on the client-side? The scenario you're describing relies on the code being executed in the client-side part of your Vercel app.

<!-- gh-comment-id:2003647701 --> @BruceMacD commented on GitHub (Mar 18, 2024): @jakobhoeg could this code be executing on the server-side rather than on the client-side? The scenario you're describing relies on the code being executed in the client-side part of your Vercel app.
GiteaMirror commented 2026-04-12 11:22:29 -05:00
Author
Owner
Copy Link

@jakobhoeg commented on GitHub (Mar 18, 2024):

@BruceMacD I am using Vercels useChat() from their @ai/react package on the client side to call the api route if that helps?

<!-- gh-comment-id:2003696230 --> @jakobhoeg commented on GitHub (Mar 18, 2024): @BruceMacD I am using Vercels `useChat()` from their `@ai/react` package on the client side to call the api route if that helps?
GiteaMirror commented 2026-04-12 11:22:30 -05:00
Author
Owner
Copy Link

@jakobhoeg commented on GitHub (Mar 18, 2024):

I also tried using the Ollama OpenAI completions instead and followed the guide found here (the Vercel AI SDK part). So my API route (/api/chat/route.ts) looks like this:

import OpenAI from 'openai';
import { OpenAIStream, StreamingTextResponse } from 'ai';

// Create an OpenAI API client (that's edge friendly!)
const openai = new OpenAI({
  baseURL: 'http://localhost:11434/v1',
  apiKey: 'ollama',
});

// IMPORTANT! Set the runtime to edge
export const runtime = 'edge';

export async function POST(req: Request) {
  const { messages, selectedModel } = await req.json();

  try {
    const response = await openai.chat.completions.create({
      model: 'gemma:2b',
      stream: true,
      messages,
    });

    // Convert the response into a friendly text-stream
  const stream = OpenAIStream(response);

  // Respond with the stream
  return new StreamingTextResponse(stream);
  } catch (error) {
    console.error('Error:', error);
  }

}

When doing so, I get these 2 errors in my Vercel logs:

Error: 403 error code: 1003
    at (node_modules/openai/error.mjs:46:19)
    at (node_modules/openai/core.mjs:256:24)
    at (node_modules/openai/core.mjs:299:29)
    at (src/app/api/chat/route.ts:17:21)
    at (node_modules/next/dist/esm/server/future/route-modules/app-route/module.js:189:36)
    at (node_modules/next/dist/esm/server/future/route-modules/app-route/module.js:128:25)
    at (node_modules/next/dist/esm/server/future/route-modules/app-route/module.js:251:29)
    at (node_modules/next/dist/esm/server/web/edge-route-module-wrapper.js:81:20)
    at (node_modules/next/dist/esm/server/web/adapter.js:157:15) {
  status: 403,
  headers: {
  cache-control: 'private, max-age=0, no-store, no-cache, must-revalidate, post-check=0, pre-check=0',
  cf-connection-close: 'close',
  connection: 'keep-alive',
  content-length: '16',
  content-type: 'text/plain; charset=UTF-8',
  date: 'Mon, 18 Mar 2024 15:00:45 GMT',
  expires: 'Thu, 01 Jan 1970 00:00:01 GMT',
  referrer-policy: 'same-origin',
  x-frame-options: 'SAMEORIGIN'
},
  error: undefined,
  code: undefined,
  param: undefined,
  type: undefined
}

AND this:

[POST] /api/chat reason=EDGE_FUNCTION_INVOCATION_FAILED, status=500, user_error=true - this one actually shows as a 405 eror on Vercel, despite saying status=500.

<!-- gh-comment-id:2004165015 --> @jakobhoeg commented on GitHub (Mar 18, 2024): I also tried using the Ollama OpenAI completions instead and followed the guide found [here](https://ollama.com/blog/openai-compatibility) (the Vercel AI SDK part). So my API route (/api/chat/route.ts) looks like this: ``` import OpenAI from 'openai'; import { OpenAIStream, StreamingTextResponse } from 'ai'; // Create an OpenAI API client (that's edge friendly!) const openai = new OpenAI({ baseURL: 'http://localhost:11434/v1', apiKey: 'ollama', }); // IMPORTANT! Set the runtime to edge export const runtime = 'edge'; export async function POST(req: Request) { const { messages, selectedModel } = await req.json(); try { const response = await openai.chat.completions.create({ model: 'gemma:2b', stream: true, messages, }); // Convert the response into a friendly text-stream const stream = OpenAIStream(response); // Respond with the stream return new StreamingTextResponse(stream); } catch (error) { console.error('Error:', error); } } ``` When doing so, I get these 2 errors in my Vercel logs: ``` Error: 403 error code: 1003 at (node_modules/openai/error.mjs:46:19) at (node_modules/openai/core.mjs:256:24) at (node_modules/openai/core.mjs:299:29) at (src/app/api/chat/route.ts:17:21) at (node_modules/next/dist/esm/server/future/route-modules/app-route/module.js:189:36) at (node_modules/next/dist/esm/server/future/route-modules/app-route/module.js:128:25) at (node_modules/next/dist/esm/server/future/route-modules/app-route/module.js:251:29) at (node_modules/next/dist/esm/server/web/edge-route-module-wrapper.js:81:20) at (node_modules/next/dist/esm/server/web/adapter.js:157:15) { status: 403, headers: { cache-control: 'private, max-age=0, no-store, no-cache, must-revalidate, post-check=0, pre-check=0', cf-connection-close: 'close', connection: 'keep-alive', content-length: '16', content-type: 'text/plain; charset=UTF-8', date: 'Mon, 18 Mar 2024 15:00:45 GMT', expires: 'Thu, 01 Jan 1970 00:00:01 GMT', referrer-policy: 'same-origin', x-frame-options: 'SAMEORIGIN' }, error: undefined, code: undefined, param: undefined, type: undefined } ``` AND this: `[POST] /api/chat reason=EDGE_FUNCTION_INVOCATION_FAILED, status=500, user_error=true` - this one actually shows as a 405 eror on Vercel, despite saying `status=500`.
GiteaMirror commented 2026-04-12 11:22:30 -05:00
Author
Owner
Copy Link

@rajasegar commented on GitHub (Sep 28, 2024):

This one worked for me:

import { Ollama } from '@langchain/ollama'

const llm = new Ollama({
  model: 'tinyllama',
  temperature: 0,
  maxRetries: 2,
  baseUrl: "http://127.0.0.1:11434"
});

const inputText = "why is the sky blue?";
const completion = await llm.invoke(inputText);
console.log(completion)

I am using Raspberry Pi 5 with Node.js 18.19.0 and ollama 0.3.10
this is my package.json

{
  "name": "langchain-tutorial",
  "version": "1.0.0",
  "description": "",
  "main": "index.js",
  "scripts": {
    "test": "echo \"Error: no test specified\" && exit 1"
  },
  "type": "module",
  "keywords": [],
  "author": "",
  "license": "ISC",
  "dependencies": {
    "@langchain/core": "^0.3.3",
    "@langchain/ollama": "^0.1.0"
  }
}
<!-- gh-comment-id:2380647094 --> @rajasegar commented on GitHub (Sep 28, 2024): This one worked for me: ```js import { Ollama } from '@langchain/ollama' const llm = new Ollama({ model: 'tinyllama', temperature: 0, maxRetries: 2, baseUrl: "http://127.0.0.1:11434" }); const inputText = "why is the sky blue?"; const completion = await llm.invoke(inputText); console.log(completion) ``` I am using Raspberry Pi 5 with Node.js 18.19.0 and ollama 0.3.10 this is my package.json ```json { "name": "langchain-tutorial", "version": "1.0.0", "description": "", "main": "index.js", "scripts": { "test": "echo \"Error: no test specified\" && exit 1" }, "type": "module", "keywords": [], "author": "", "license": "ISC", "dependencies": { "@langchain/core": "^0.3.3", "@langchain/ollama": "^0.1.0" } } ```
GiteaMirror commented 2026-04-12 11:22:30 -05:00
Author
Owner
Copy Link

@titiyoyo commented on GitHub (Oct 27, 2024):

@rajasegar worked for me too, thanks for posting this !

<!-- gh-comment-id:2439916452 --> @titiyoyo commented on GitHub (Oct 27, 2024): @rajasegar worked for me too, thanks for posting this !
GiteaMirror commented 2026-04-12 11:22:31 -05:00
Author
Owner
Copy Link

@rosariogueli commented on GitHub (Oct 30, 2024):

This worked for me, using OpenAI nodejs SDK, hope that helps!

import OpenAIApi from 'openai';

const openai = new OpenAIApi({
  apiKey: 'nope',
  baseURL: 'http://127.0.0.1:11434/v1',
});

try {
  const response = await openai.chat.completions.create({
    model: 'llama3',
    messages: [
      { role: 'user', content: 'Hello, how are you?' }
    ],
  });
  console.log(response?.choices?.[0]?.message?.content)
} catch (err) {
  console.log(err.message);
}
<!-- gh-comment-id:2445588084 --> @rosariogueli commented on GitHub (Oct 30, 2024): This worked for me, using OpenAI nodejs SDK, hope that helps! ``` import OpenAIApi from 'openai'; const openai = new OpenAIApi({ apiKey: 'nope', baseURL: 'http://127.0.0.1:11434/v1', }); try { const response = await openai.chat.completions.create({ model: 'llama3', messages: [ { role: 'user', content: 'Hello, how are you?' } ], }); console.log(response?.choices?.[0]?.message?.content) } catch (err) { console.log(err.message); } ```
GiteaMirror commented 2026-04-12 11:22:31 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Dec 19, 2024):

This issue has gone pretty stale. I'm going to go ahead and close it, but I'm curious is there was any resolution.

<!-- gh-comment-id:2555813180 --> @pdevine commented on GitHub (Dec 19, 2024): This issue has gone pretty stale. I'm going to go ahead and close it, but I'm curious is there was any resolution.
GiteaMirror commented 2026-04-12 11:22:31 -05:00
Author
Owner
Copy Link

@dungdo123 commented on GitHub (Feb 26, 2025):

I run n8n and ollama on windows, set the base URL: http://172.21.0.1:11434/ fixed the ECONNREFUSED problem.

<!-- gh-comment-id:2683743450 --> @dungdo123 commented on GitHub (Feb 26, 2025): I run n8n and ollama on windows, set the base URL: http://172.21.0.1:11434/ fixed the ECONNREFUSED problem.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug js needs more info
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#1468
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?