Function request_claude_token

src/gpt_request.py:72–121

(prompt, log_id=None, max_tokens=10000, max_retries=3)

def request_claude_token(prompt, log_id=None, max_tokens=10000, max_retries=3):
    base_url = cfg("claude", "base_url")
    api_key = cfg("claude", "api_key")
    client = OpenAI(base_url=base_url, api_key=api_key)

    if log_id is None:
        log_id = generate_log_id()

    extra_headers = {"X-TT-LOGID": log_id}
    usage_info = {"prompt_tokens": 0, "completion_tokens": 0, "total_tokens": 0}

    retry_count = 0
    while retry_count < max_retries:
        try:
            completion = client.chat.completions.create(
                model="claude-4-opus",
                messages=[
                    {
                        "role": "user",
                        "content": [
                            {
                                "type": "text",
                                "text": prompt,
                            },
                        ],
                    }
                ],
                max_tokens=max_tokens,
                extra_headers=extra_headers,
            )
            # --- MODIFIED: token usage ---
            if completion.usage:
                usage_info["prompt_tokens"] = completion.usage.prompt_tokens
                usage_info["completion_tokens"] = completion.usage.completion_tokens
                usage_info["total_tokens"] = completion.usage.total_tokens
            return completion, usage_info

        except Exception as e:
            retry_count += 1
            if retry_count >= max_retries:
                raise Exception(f"Failed after {max_retries} attempts. Last error: {str(e)}")

            # Exponential backoff with jitter
            delay = (2**retry_count) * 0.1 + (random.random() * 0.1)
            print(
                f"Request failed with error: {str(e)}. Retrying in {delay:.2f} seconds... (Attempt {retry_count}/{max_retries})"
            )
            time.sleep(delay)

    return None, usage_info


def request_gemini_with_video(prompt: str, video_path: str, log_id=None, max_tokens: int = 10000, max_retries: int = 3):