Synara v0.8.4 发布：默认 Codex 模型换成 GPT-6 Astra
这个开发工具更新 66 个 PR，把 GPT-6 Astra 设为 Codex 默认模型，并称本地负载内存降 55%、启动快 45%。
66 PRs later, Synara v0.8.4 is finally live This update turned into a major overhaul of the entire app: → Edit files and Git diffs directly inside Synara → More powerful Git reviews with blame, word-level diffs, and branch comparisons → Better browser automation with saved sessions and protected logins → A new effort slider and redesigned Add panel → Guided onboarding for new users → GPT-6 Astra as the default Codex model → More reliable providers, transcripts, simulators, and PR workflows → Broad performance and visual improvements across the app In focused local workloads, we measured up to 55% less temporary memory usage, 35.8% less Electron CPU usage, and up to 45.2% faster warm chat opening This one feels like a completely different, better Synara, and this is nothing yet
这条是引用着另一条写的：「Synara v0.8.4 把 GPT-6 Astra 设为默认 Codex 模型」