What causes "Invalid model selection" in Mintplex-Labs/anything-llm?

Select a concrete model for the chosen provider before saving

Mintplex-Labs/anything-llm · warning · Error

Invalid model selection

Error message

Invalid model selection

What it means

Companion guard for non-router providers: validatedModelSelection(selectedLLMModel) must yield a usable model id, and a falsy result throws the localized 'model-router.chat.invalid-model' message ('Invalid model selection') before Workspace.update is called. It catches the case where the provider is chosen but no concrete model is set.

Source

Thrown at frontend/src/components/WorkspaceChat/ChatContainer/PromptInput/LLMSelector/index.jsx:101

  async function handleSave() {
    setSaving(true);
    try {
      setHasChanges(false);

      const isRouter = selectedLLMProvider === "anythingllm-router";
      if (isRouter && !selectedRouterId)
        throw new Error(t("model-router.chat.select-router-error"));

      const updateData = isRouter
        ? { chatProvider: selectedLLMProvider, router_id: selectedRouterId }
        : {
            chatProvider: selectedLLMProvider,
            chatModel: validatedModelSelection(selectedLLMModel),
          };

      if (!isRouter && !updateData.chatModel)
        throw new Error(t("model-router.chat.invalid-model"));

      const { message } = await Workspace.update(slug, updateData);

      if (!!message) throw new Error(message);
      window.dispatchEvent(new Event(SAVE_LLM_SELECTOR_EVENT));
    } catch (error) {
      console.error(error);
      showToast(error.message, "error", { clear: true });
    } finally {
      setSaving(false);
    }
  }

  const providerName =
    WORKSPACE_LLM_PROVIDERS.find((p) => p.value === selectedLLMProvider)
      ?.name || selectedLLMProvider;

  if (loading) {

View on GitHub (pinned to 3aec848f28)

Solutions

Select a concrete model for the chosen provider before saving
Add valid credentials for the provider so its model list populates, then pick a model
For custom-model providers, type a non-empty model id in the input

Defensive patterns

Strategy: validation

Validate before calling

const model = selectedLLMProvider === "anythingllm-router" ? null : validatedModelSelection(selectedLLMModel);
const canSave = selectedLLMProvider === "anythingllm-router" ? !!selectedRouterId : !!model;
// disable Save when !canSave and flag the model dropdown

Try / catch

try { await handleSave(); }
catch (e) { if (e.message === t("model-router.chat.invalid-model")) highlightModelDropdown(); else throw e; }

Prevention

Require provider credentials to load the model list before enabling save
Normalize placeholder model values ('all', '') to null so validation fails early and visibly
Re-validate the model whenever the provider changes, not only on submit

When it happens

Trigger: A non-router provider is selected while its model value is empty, a placeholder that normalizes to null, or the model list never loaded (missing credentials) so nothing was chosen.

Common situations: Provider lacks an API key so the model dropdown stayed empty; custom-model input cleared; switching providers quickly and saving before the new model list resolves; model value filtered out by validation.

Related errors

AI-assisted analysis of Mintplex-Labs/anything-llm@3aec848f28 (2026-08-18). Data as JSON: /api/errors/be4a4fe1272e6f34. Report an issue: GitHub.