Error reading content

[Anh_Le]

Hello everyone,

I have a question. When I write code to load a model in OpenAI’s format, I encounter this error. I have tested it with OpenAI’s library, and it worked successfully.

Here is my JSON format:

{
  "id": "chatcmpl-30101561cf",
  "object": "chat.completion",
  "created": 1742132277,
  "model": "DeepSeek-R1-Distill-Qwen-7B-Q4_K_M",
  "choices": [
    {
      "index": 0,
      "logprobs": null,
      "finish_reason": "stop",
      "message": {
        "role": "assistant",
        "content": "</think>\n\n1 + 1 equals **2**."
      }
    }
  ],
  "usage": {
    "prompt_tokens": 10,
    "completion_tokens": 6,
    "total_tokens": 16
  },
  "stats": {
    "tokens_per_second": 3.394546566015665,
    "time_to_first_token": 0.111,
    "generation_time": 1.7675409317016602,
    "stop_reason": "eosFound"
  },
  "model_info": {
    "arch": "deepseek",
    "quant": "Q4_K_M",
    "format": "gguf",
    "context_length": 2048
  },
  "runtime": {
    "name": "llama.cpp",
    "version": "1.3.0",
    "supported_formats": [
      "gguf"
    ]
  }
}

Can anyone help me? Thank you!

[jksr]

Hey @Anh_Le - Could you maybe post your workflow inside code blocks like this on a new line

type or paste code here

[bartv]

I updated their post. Pro-tip: you can even add json syntax higlighting by using the following code to start your code block:

```json

[jksr]

@Anh_Le I am curious - if I get it right you want to use another endpoint but still the OpenAI API. Where does your JSON come into play? Usually you would just switch the base_url and the model. I marked the options in this screenshot.

Please clarify if I misunderstood you