✨ Feature: Add feature: Support for text-to-speech endpoint /v1/audio/speech

🐛 Bug: Fix the bug in the vertex API key concatenation syntax error.

✨ Feature: Add feature: support setting rate limit for each model individually

✨ Feature: Add support for embeddings model

✨ Feature: Add an "image" field to the configuration file to customize control over whether to enable image reading support.

✨ Feature: 1. Add support for experimental frontend.

remove unused import of logger

feat: add TextToSpeechRequest model and implement audio speech endpoint with processing logic

🐛 Bug: Fix the bug that prevents the voice function from being used.

🐛 Bug: Fix the bug where the Pillow dependency is not installed

🐛 Bug: 1. Fix the bug that causes an error when Claude uploads a PNG image.

✨ Feature: Add support for v1/moderations endpoint.

✨ Feature: Add support for v1/audio/transcriptions endpoint

🐛 Bug: Fix the bug where the o1 model cannot obtain text format.

🤖 Models: Add support for the cohere series models

🐛 Bug: Fix the bug where the lobechat gemini official API cannot use the tool. The reason is that lobechat automatically uses multiple consecutive underscores in the system prompt to name function names, which gemini does not support.

🐛 Bug: Fix the bug where the system prompt is not deleted in the o1-mini model message.

🤖 Models: Add support for o1-mini o1-preview model

🐛 Bug: Fix the bug where the client uploads the image URL instead of converting it to base64 encoding during image recognition.

🐛 Bug: Fix the bug that the official Claude API cannot use PNG format images.

✨ Feature: Add feature: Support Cloudflare API model

✨ Feature: Adapt to the deprecated function call request body format of OpenAI

🐛 Bug: Fix the bug where tool use is still included in the request body when tools are closed.

✨ Feature: Add features: Add API channel success rate statistics, channel status records.

🐛 Bug: Fix the bug of tool use request body format error

Fix the bug where claude tool_choice does not support "tool_choice": {"type": "any"}

🐛 Bug: 1. Fix the bug where the max_tokens is missing in the Claude request body

Add feature: support load balancing for multiple keys in a single channel, enabled by default.

Fix the bug where claude stream is not defined.

Fix the bug where Claude cannot use tool use.

Add feature: support OpenAI dall-e-3 image generation

Fix the bug of incorrect request body format for tool use.

fix: Improve round-robin provider selection and add Claude max_tokens auto-fill

Fix the bug where Gemini can only use the first region.

Add feature: support vertex claude API using tool use functionality.

Add Gemini region load balancing.

✨ Feature: Add feature: Support vertex ai API and vertex tool use invocation.

✨ Feature: Add feature: support Gemini API tool use

🐛 Bug: Fix the bug when the date plugin does not require parameters.

Upgrade the request body field Function, use the Tool field for function calls.

Upgrade the request body field Function, use the Tool field for function calls.

Fix the bug causing errors in the GPT channel when using the Wu API.

gemini adds support for system prompts

Fix the issue where Gemini cannot be used.

Fixed the bug in openrouter sse special format handling.

🐛 Bug: 1. Fix the bug where system_prompt is undefined

✨ Feature: Support Claude 3.5 8192 maximum output length.

Fix the error of name not defined

Determine whether to use tools through the configuration file tools.

Fix the issue where openrouter cannot use Gemini.