Cannot read properties of undefined when changing "thinking effort" on agent

[niklastanner]

What happened?

I have issues creating a second agent with different settings.
When setting the thinking-effort to high I cannot use the agent anymore. I always get errors, even if I change the thinking-effort setting back to the previously working value. The only solution to fix this I've found so far is to remove the agent and recreate it from scratch.

My settings

I use an Azure OpenAI Endpoint provided using Azure AI Foundry (https://<INSTANCE>.cognitiveservices.azure.com/openai/deployments/<DEPLOYMENT>/chat/completions?api-version=2025-01-01-preview). The issue occurs on all models.

Version Information

Helm chart version: 1.8.10
Docker image version: ghcr.io/danny-avila/librechat:v0.7.7

Steps to Reproduce

1. Create Agent "A"
2. Create Agent "B" with same configuration but different name. Set the thinking-effort to high
3. Start conversation and ask anything

What browsers are you seeing the problem on?

No response

Relevant log output

{"level":"error","message":"[handleAbortError] AI response error; aborting request: Cannot read properties of undefined (reading 'replace')","stack":"TypeError: Cannot read properties of undefined (reading 'replace')\n    at sanitizeModelName (/app/api/utils/azureUtils.js:10:20)\n    at getLLMConfig (/app/api/server/services/Endpoints/openAI/llm.js:99:9)\n    at initializeClient (/app/api/server/services/Endpoints/openAI/initialize.js:145:21)\n    at initializeAgentOptions (/app/api/server/services/Endpoints/agents/initialize.js:128:25)\n    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)\n    at async initializeClient (/app/api/server/services/Endpoints/agents/initialize.js:208:25)\n    at async AgentController (/app/api/server/controllers/agents/request.js:47:24)\n    at async /app/api/server/routes/agents/chat.js:37:5"}

Screenshots

No response

Code of Conduct

• I agree to follow this project's Code of Conduct

[danny-avila]

No issue here:

Furthermore, you can click on "Reset Model Parameters" to reset all parameters then "Save" when editing an existing agent.

Make sure you are using the latest version:

Updating Instructions (docker):
https://www.librechat.ai/docs/local/docker#update-librechat

[niklastanner]

I can confirm that updating to v0.7.8 does solve the issue. Unfortunately the Helm chart does use 0.7.7 and not the official docker image, even if it is referenced by the official documentation. But as far as I understand the chart will be migrated into the official repo somehow. So I'll just stick with a manual tagging for now.