|
|
c9c3eedd82
|
feat: 移除实验性间隔思考模式,推理强度前置到前端
- 彻底移除「每 N 次请求强制思考」的间隔调度机制(思考/非思考交替导致缓存命中率极低)
- 运行模式收敛为 fast/thinking:原 deep 改名 thinking,历史数据读取时映射
- 新增推理强度五档滑块(low/medium/high/xhigh/max + 默认不传参),置于模型/模式弹窗
- 模型配置新增 reasoning_effort 开关;档位会话级持久化到对话 meta,随对话保存/恢复
- 个人空间新增「默认推理强度」;新用户默认运行模式改为 thinking
- 滑块交互:JS 插值驱动(点击/拖动/吸附全程平滑),首帧档位语义定位无闪烁
- 保存链路:前端防抖 600ms + conversation_id 兜底,修复 /new 发消息丢档位、切对话恢复错误
|
2026-07-21 13:47:53 +08:00 |
|
|
|
d541d9bd30
|
fix: extend run_command foreground timeout limit to 120s
|
2026-04-25 16:34:56 +08:00 |
|
|
|
e814e89e32
|
feat: add conversation-bound permission modes and tool approval flow
|
2026-04-11 04:07:28 +08:00 |
|
|
|
b63e827ed2
|
fix: increase tool output char limits to 50000
|
2026-04-08 16:52:39 +08:00 |
|
|
|
877bcc2fad
|
fix: improve api error diagnostics and raise model quotas
|
2026-03-06 17:02:19 +08:00 |
|
|
|
b81d6760bd
|
feat: containerize terminals and add resource controls
|
2025-11-23 18:49:35 +08:00 |
|
|
|
aee18837e4
|
feat: enhance smart thinking scheduling
|
2025-11-20 17:25:30 +08:00 |
|
|
|
dacc68f46a
|
feat: enhance read tool and config structure
|
2025-11-14 18:33:55 +08:00 |
|