Asep Haryana
1ae8a53cd4
chore: ignore Nix build result symlink
2026-07-30 18:33:52 +07:00
Asep Haryana
ff52f8841c
feat(infra): Nix build for hub app, systemd deployment
...
- Add flake.nix with derivations for hub (Next.js), scraper, tools, llm-api
- Create hub systemd unit (port 3099)
- Update Traefik dynamic config to point to host hub service
- Add iptables rule for Docker-to-host communication
- Use fetchGit for submodule source resolution
2026-07-30 18:33:43 +07:00
asepharyana
be7348a27d
fix(infra): remove buffer/compress from llm-api for real SSE streaming
...
- llm-api uses llm-chain instead of common-chain
- llm-chain excludes buffer & compress middlewares
- Prevents Traefik from buffering streaming SSE responses
- Each token is now flushed immediately
2026-07-26 19:14:52 +07:00
asepharyana
43f0df4493
fix(infra): remove buffer/compress from llm-api middleware chain
...
- Buffer middleware causes SSE stream buffering
- Compress middleware breaks real-time streaming
- Added llm-chain with only secure-headers + retry + rate-limit
- llm-api router now uses llm-chain@file instead of common-chain@file
2026-07-26 19:13:08 +07:00
asepharyana
a4687c9d5b
fix(llm-api): update submodule - reasoning/content stream separation
2026-07-26 19:04:36 +07:00
asepharyana
d99bc06df3
fix(llm-api): update submodule to 6ff31b5 — Jinja template via minijinja
...
- Replaced llama-cpp-2 apply_chat_template with minijinja rendering
- reasoning_content separate from content in API response
- Works with MiniCPM5 thinking model natively
2026-07-26 18:21:43 +07:00
asepharyana
10aeda4d4b
fix: update llm-api MODEL_PATH for new MiniCPM5 GGUF
...
Updates MODEL_PATH to MiniCPM5-1B-Claude-Opus-Fable5-V2-Thinking-Q8_0.gguf
2026-07-26 15:40:14 +07:00
asepharyana
05d6b4a055
chore: update llm-api submodule to e1f5195 (SSE streaming)
2026-07-25 11:34:30 +07:00
asepharyana
8fc4a15e0c
feat(llm-api): add MiniCPM-V LLM API service with Traefik routing
2026-07-25 11:20:15 +07:00
Asepharyana
1e10973c86
ci: trigger fresh run for updated security.yml
2026-07-24 01:39:57 +07:00
Asepharyana
b7e4203a15
fix(security): checkout git submodules for CodeQL Rust analysis
2026-07-24 01:36:17 +07:00
asepharyana
45ab3c8792
test: pr-agent v3 with model fix
2026-07-17 08:03:26 +07:00
Asep Haryana