Hello @Vapi ! I would like to know if deepseek models will be available to use for voice assistants. Currently, I see the v3 and r1 models, but when I select them and try to make a call it fails with error: call.in-progress.error-providerfault-deep-seek-llm-failed. Do i need to set something else in VAPI to use them? Here are the failing calls: https://dashboard.vapi.ai/calls/da2f7b01-b894-4152-8ada-db1679ac2df6 and https://dashboard.vapi.ai/calls/91b2b7f7-d660-49b0-aecb-09ecceb121a7 Thank you very much!

Thank you for sharing the error details. To help diagnose the issue with the Deepseek v3 and r1 models, could you confirm if you have tried using the new

deepseek-reasoner

model or the

deepseek-r1-distill-llama-70b

(with Groq) as mentioned in the latest Vapi documentation? This will help determine if the problem is specific to certain model versions or a broader provider issue. Error code:

call.in-progress.error-providerfault-deep-seek-llm-failed

Reference: [Vapi call end reasons](https://docs.vapi.ai/calls/call-ended-reason) Let me know your results, and we can proceed from there! Sources: - [Call end reasons](https://docs.vapi.ai/calls/call-ended-reason) - [Changelog January 22, 2025](https://docs.vapi.ai/changelog/2025/1/22) - [Changelog February 1, 2025](https://docs.vapi.ai/changelog/2025/2/1)

I send a screen shot of how i am selecting the model: Provider: Deep seek, Model: DeepSeek-V3 https://cdn.discordapp.com/attachments/1412847388439347230/1412853728414925022/image.png?ex=68b9ce12&is=68b87c92&hm=e174db7699169f93cfbe5495711e63e366bd6e4ccc6cda4f892410c03193d173&

Hey! To help track down this issue, could you share: \- The call ID \- When exactly this happened (the timestamp) \- What response you expected to get \- What response you actually got instead This would really help us figure out what went wrong!

Here are the failing calls: - https://dashboard.vapi.ai/calls/da2f7b01-b894-4152-8ada-db1679ac2df6 - TS: 1756916419044 - I expected a response from the DeepSeek-V3 model - call.in-progress.error-providerfault-deep-seek-llm-failed - https://dashboard.vapi.ai/calls/91b2b7f7-d660-49b0-aecb-09ecceb121a7 - TS: 1756916448430 - Same as above

Hey Jeronimo, thanks for sharing your details. Our team will look into it and get back to you soon with an update.

Super! Thank you Preveen!

Hey, here’s the info on what's happening: Root cause: DeepSeek API returned 402 “Insufficient Balance” for your BYOK key. Evidence: “Model request failed … Insufficient Balance”; endedReason “call.in-progress.error-providerfault-deep-seek-llm-failed”; 0 tokens billed. No fallback model was configured, so the pipeline aborted immediately. Fix: add credits to DeepSeek or configure a fallback (e.g., OpenAI/Claude). I didn’t run any tools automatically. Let me know if you need anything else.