docs: add technical report for codex handoff (2026-06-13)

fix: remove --voxcpm2-base-lm/acoustic args (not valid for llama-server CLI)

test_omni: add stderr capture on startup failure

fix: rewrite test_omni to use llama-server built-in omni endpoints

refactor: deprecate llama-omni-server, use llama-server built-in omni endpoints

fix: set n_parallel=1, n_ctx=4096 before omni_init to prevent n_seq_max > 256

test_omni: add stderr capture for omni-server when init fails

fix: params.model is common_params_model struct, use .path

tmp: health check / omni test scripts for debugging omni_init 500

fix: set params.model/vpm/apm/tts in server-omni.cpp omni_init handler so model files are found

modal/omni: add llama-omni-server for full-duplex voice API

Fix Gradio duplex audio playback clearing

Stabilize Gradio duplex voice flow

modal/omni: PRODUCTION DEPLOY - text inference fully verified, llama-server running on T4

modal/omni: text inference verified (Chinese OK, English needs fix)

modal/omni: llama-server inference verified on T4

docs: add production script protect warnings to deploy guide

docs: comprehensive Modal deploy guide (both solutions + pitfalls)

chore: add build_llama_server.sh (local cross-compile script, unused)

deploy_omni: llFile llama.cpp-omni compile succeeds on Modal T4

清理临时文件

[cleanup] 添加临时文件 gitignore + 删除 _fix_*.py 临时修复脚本

[PregoPal] 2026-06-11 全双工语音对话开发阶段完成

docs: 更新开发日志和本地部署经验，记录全双工实现细节

Merge branch 'main' of https://huggingface.co/spaces/build-small-hackathon/PregoPal

Phase 6: 端到端测试通过 + go_server bugfix

Merge branch 'main' of https://huggingface.co/spaces/build-small-hackathon/PregoPal

用户手动备份

Merge branch 'main' of https://huggingface.co/spaces/build-small-hackathon/PregoPal

用户手动备份

用户手动备份

全双工语音对话实现

技术路径

用户手动修改

用户手动整理文件，并备份deploy_backup.py文件

merge: 解决合并冲突

refactor: 将三天总结从框架层面合并到营养报告，统一UI设计

docs: 全面更新README，同步最新UI架构与部署状态

docs: 部署成功报告 + 清理神秘文件 c

core: 更新 ModelLoader 对接远端 Modal API + README 同步

chore: 清理临时文档(tmp_*)文件

modal: 降级 GPU A100 -> T4 节省成本

Merge branch 'main' of https://huggingface.co/spaces/build-small-hackathon/PregoPal

docs: 补充两种方案对比(wheel vs 源码编译)，明确预编译wheel仍属llama.cpp

docs: 技术日志-解释为什么预编译wheel部署这么快

deploy: 成功部署 MiniCPM-o-4_5 到 Modal (预编译llama-cpp-python A100)

docs: pre-refactor backup - before tech log integration

fix: Key Nutrients营养素名双语 + Persistent Deficits紧凑标签式 + Food Suggestions扩大展示(1:2) + Summary结构化UI卡片

fix(deploy): 使用预编译CUDA wheel替代源码编译

fix: 三天总结营养素名中英文双语显示 + CSS消除灰色填充和双重方框