v0.9.10

What's Changed

• feat: 新增支持豆包语音合成2.0功能 by @feitianbubu in #2067
• fix: gemini batch embedding token not counted by @creamlike1024 in #2061
• fix: handle JSON parsing for thinking content in ollama stream by @somnifex in #2065
• feat: 豆包语音2.0音色支持情感,情绪,音量 by @feitianbubu in #2068

Full Changelog: v0.9.9...v0.9.10

v0.9.9-patch.4

feat: 防呆优化，允许OpenAI渠道调用sora异步任务

v0.9.9-patch.3

feat: 添加sora模型仅按次计费的解决方案 (Add pay-per-use solution for the Sora model)

如何使用 How to Use

环境变量设置 TASK_PRICE_PATCH=sora-2,sora-2-pro
对所有task任务都有效，模型名用英文逗号,隔开
设置后，任务将不会按时长，大小等自动计算

Set the environment variable TASK_PRICE_PATCH=sora-2,sora-2-pro
This applies to all task operations, with model names separated by English commas ,.

v0.9.9-patch.2

fix: 修复/v1/videos失败不退费的问题

v0.9.9-patch.1

fix: 修复gemini embedding无法使用的问题

v0.9.9

feat: 修复aws渠道-thinking后缀不生效的问题
feat: 支持aws渠道使用api key
feat: 支持aws渠道使用链接媒体

v0.9.8-patch.1

feat: sora任务优化

v0.9.8

What's Changed

• feat: jimeng use openai sdk input_reference i2v by @feitianbubu in #2029
• fix: 修复视频任务不同分组可能导致补回额度计算错误 by @xyfacai in #2030
• feat: jimeng images base64 limit by @feitianbubu in #2032
• fix(convert): 修复 OpenAI 转 Claude 流时 thinking 块的格式问题 by @Inblac in #2035
• feat: 添加SiliconFlow图像生成接口自动转换支持 by @etnAtker in #2036
• feat: endpoint type log by @seefs001 in #2038
• feat: 支持谷歌分析和Umami分析

谷歌分析设置文档

https://docs.newapi.pro/wiki/analytics-setup/

New Contributors

• @Inblac made their first contribution in #2035
• @etnAtker made their first contribution in #2036

Full Changelog: v0.9.7...v0.9.8

v0.9.7-patch.2

What's Changed

• fix: empty version by @seefs001 in #2024
• feat: Add pre-status protection for IncreaseUserQuota by @feitianbubu in #2025
• refactor: Openai video model 移动到 dto by @xyfacai in #2027
• feat: jimeng use openai sdk input_reference i2v by @feitianbubu in #2029
• fix: 修复视频任务不同分组可能导致补回额度计算错误 by @xyfacai in #2030
• feat: jimeng images base64 limit by @feitianbubu in #2032

Full Changelog: v0.9.7-patch.1...v0.9.7-patch.2

v0.9.7-patch.1

What's Changed

• fix: ignore ghcr by @seefs001 in #2022
• fix: version by @seefs001 in #2023
• feat: feat: support free model setting （支持免费模型功能）

Full Changelog: v0.9.7...v0.9.7-patch.1