gramps/jarvisChat
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fix: restore EMBED_URL pointing to ollama on 192.168.50.210:11434

Browse Source
This commit is contained in:
gramps
2026-06-27 16:03:19 -07:00
parent 4b36fd315a
commit 66b086c3f3
3 changed files with 5 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
AGENTS.md
View File

@@ -102,7 +102,7 @@ The upstream request includes `"logprobs": true`. `parse_llama_stream_chunk()` e
- `ALLOWED_SETTINGS_KEYS` in `config.py` controls which keys the UI can write via `/api/settings` - `ALLOWED_SETTINGS_KEYS` in `config.py` controls which keys the UI can write via `/api/settings`
- Settings table seeded with defaults (`profile_enabled`, `search_enabled`, `memory_enabled`, `skills_enabled`, `default_model`) — never overwritten by `init_db()` - Settings table seeded with defaults (`profile_enabled`, `search_enabled`, `memory_enabled`, `skills_enabled`, `default_model`) — never overwritten by `init_db()`
- Profile table uses singleton row `id=1` - Profile table uses singleton row `id=1`
- RAG embedding requests go to `LLAMA_SERVER_BASE` at `/api/embeddings` - RAG embedding requests go to `EMBED_URL` at `/api/embeddings` (separate Ollama instance)
### SSE Protocol ### SSE Protocol

2
README.md
View File

@@ -12,7 +12,7 @@ Developer wiki: [docs/wiki/Home.md](docs/wiki/Home.md)
- **`COMPLETIONS_API_KEY`** — auto-generated secret key for the OpenAI-compatible endpoint, overridable via `JARVISCHAT_COMPLETIONS_API_KEY` env var - **`COMPLETIONS_API_KEY`** — auto-generated secret key for the OpenAI-compatible endpoint, overridable via `JARVISCHAT_COMPLETIONS_API_KEY` env var
- **Perplexity auto-search fixed** — upstream request now sends `"logprobs": true`, `parse_llama_stream_chunk()` extracts per-token logprobs, so `calculate_perplexity()` and `is_uncertain()` work correctly (was dead code) - **Perplexity auto-search fixed** — upstream request now sends `"logprobs": true`, `parse_llama_stream_chunk()` extracts per-token logprobs, so `calculate_perplexity()` and `is_uncertain()` work correctly (was dead code)
- **All `/api/models` endpoints** — now correctly target `LLAMA_SERVER_BASE` (llama-server on port 8081) instead of the old Ollama port; `/api/ps` uses `/v1/models` endpoint - **All `/api/models` endpoints** — now correctly target `LLAMA_SERVER_BASE` (llama-server on port 8081) instead of the old Ollama port; `/api/ps` uses `/v1/models` endpoint
- **RAG embedding endpoint fixed** — hardcoded `EMBED_URL` replaced with `LLAMA_SERVER_BASE` from config, respecting the `JARVISCHAT_LLAMA_SERVER_BASE` env var - **RAG embedding endpoint fixed** — `EMBED_URL` changed from old server `:8081` to correct host/port `http://192.168.50.210:11434` (Ollama on new machine)
- **Error messages corrected** — all user-facing errors say "inference server" instead of "Ollama" or "llama-server" - **Error messages corrected** — all user-facing errors say "inference server" instead of "Ollama" or "llama-server"
- **Secure SSE protocol** — raw search results are no longer leaked in the SSE event stream - **Secure SSE protocol** — raw search results are no longer leaked in the SSE event stream
- **FTS5 query safety** — operator keywords (`AND`, `OR`, `NOT`, `NEAR`) are double-quoted to prevent parse errors - **FTS5 query safety** — operator keywords (`AND`, `OR`, `NOT`, `NEAR`) are double-quoted to prevent parse errors

5
rag.py
View File

@@ -7,11 +7,12 @@ import httpx
from db import get_db, get_setting, list_skills_with_state, format_active_skills_prompt from db import get_db, get_setting, list_skills_with_state, format_active_skills_prompt
from memory import search_memories from memory import search_memories
from config import LLAMA_SERVER_BASE, MAX_SKILL_PROMPT_CHARS from config import MAX_SKILL_PROMPT_CHARS
log = logging.getLogger("jarvischat") log = logging.getLogger("jarvischat")
QDRANT_URL = "http://192.168.50.108:6333" QDRANT_URL = "http://192.168.50.108:6333"
EMBED_URL = "http://192.168.50.210:11434"
EMBED_MODEL = "mxbai-embed-large" EMBED_MODEL = "mxbai-embed-large"
RAG_COLLECTION = "jarvis_rag" RAG_COLLECTION = "jarvis_rag"
RAG_SCORE_THRESHOLD = 0.25 RAG_SCORE_THRESHOLD = 0.25
@@ -21,7 +22,7 @@ async def query_rag(query: str, limit: int = 3) -> list:
try: try:
async with httpx.AsyncClient() as client: async with httpx.AsyncClient() as client:
embed_resp = await client.post( embed_resp = await client.post(
f"{LLAMA_SERVER_BASE}/api/embeddings", f"{EMBED_URL}/api/embeddings",
json={"model": EMBED_MODEL, "prompt": query}, json={"model": EMBED_MODEL, "prompt": query},
timeout=10.0, timeout=10.0,
) )