Mintplex-Labs/anything-llm · critical · Error

KoboldCPP must have a valid base path to use for the api.

Error message

KoboldCPP must have a valid base path to use for the api.

What it means

Thrown by the KoboldCPPLLM constructor when process.env.KOBOLD_CPP_BASE_PATH is unset. KoboldCPP exposes an OpenAI-compatible endpoint when run with its --openai compatible mode; AnythingLLM needs that URL as the OpenAI client baseURL. Without it the provider cannot target the local KoboldCPP server.

Source

Thrown at server/utils/AiProviders/koboldCPP/index.js:16

const { NativeEmbedder } = require("../../EmbeddingEngines/native");
const {
  clientAbortedHandler,
  writeResponseChunk,
  formatChatHistory,
} = require("../../helpers/chat/responses");
const {
  LLMPerformanceMonitor,
} = require("../../helpers/chat/LLMPerformanceMonitor");
const { v4: uuidv4 } = require("uuid");

class KoboldCPPLLM {
  constructor(embedder = null, modelPreference = null) {
    const { OpenAI: OpenAIApi } = require("openai");
    if (!process.env.KOBOLD_CPP_BASE_PATH)
      throw new Error(
        "KoboldCPP must have a valid base path to use for the api."
      );

    this.className = "KoboldCPPLLM";
    this.basePath = process.env.KOBOLD_CPP_BASE_PATH;
    this.openai = new OpenAIApi({
      baseURL: this.basePath,
      apiKey: null,
    });
    this.model = modelPreference ?? process.env.KOBOLD_CPP_MODEL_PREF ?? null;
    if (!this.model) throw new Error("KoboldCPP must have a valid model set.");
    this.limits = {
      history: this.promptWindowLimit() * 0.15,
      system: this.promptWindowLimit() * 0.15,
      user: this.promptWindowLimit() * 0.7,
    };

    this.embedder = embedder ?? new NativeEmbedder();

View on GitHub (pinned to 526360e320)

Solutions

1. Start KoboldCPP with OpenAI-compatible mode and set KOBOLD_CPP_BASE_PATH to its URL, e.g. http://localhost:5001/v1
2. Confirm KoboldCPP is listening with curl $KOBOLD_CPP_BASE_PATH/models
3. Restart AnythingLLM after editing .env
4. Ensure the URL includes the version path the OpenAI client expects

Example fix

// before
// .env: KOBOLD_CPP_MODEL_PREF=<loaded model>   // no base path

// after
// .env:
//   KOBOLD_CPP_BASE_PATH=http://localhost:5001/v1
//   KOBOLD_CPP_MODEL_PREF=<loaded model>

function hasKoboldBasePath() {
  try {
    const u = new URL(process.env.KOBOLD_CPP_BASE_PATH);
    return u.protocol.startsWith('http');
  } catch { return false; }
}
if (!hasKoboldBasePath()) {
  throw new ConfigError('Start KoboldCPP in OpenAI mode and set KOBOLD_CPP_BASE_PATH.');
}

When it happens

Trigger: Constructing `new KoboldCPPLLM(embedder, modelPreference)` while KOBOLD_CPP_BASE_PATH is absent. Fires at provider selection when the user picked KoboldCPP but supplied no endpoint.

Common situations: KoboldCPP started without the --openai flag (so no compatible endpoint); .env missing the line; wrong port in the URL; KoboldCPP process not running; path missing the /v1 suffix.

Related errors

• No Foundry Base Path was set.
• No Gemini API key was set.
• GenericOpenAI must have a valid base path to use for the api
• No Gitee AI API key was set.
• No Groq API key was set.

AI-assisted analysis of Mintplex-Labs/anything-llm@526360e320 (2026-08-13). Data as JSON: /api/errors/d0a829932f0e724d. Report an issue: GitHub.