本地LLM编排器

# Install dependencies (requires Python 3.12+)
uv sync

# Pull required Ollama models
ollama pull gemma4:26b           # Chat model (~17GB)
ollama pull qwen3-embedding:0.6b # Embedding model (needed if RAG is re-enabled)

# Configure TSS server connection (set to the IP/port of the running TSS instance)
# Edit .env:
#   TSS_UDP_HOST=<TSS server IP>
#   TSS_UDP_PORT=14141

# Run the orchestrator
uv run python main.py

// Request
{
  "messages": [
    {"role": "user", "content": "What is the EVA1 oxygen status?"}
  ],
  "stream": false
}

// Response
{
  "role": "assistant",
  "content": "EVA1: O2 94%, CO2 0.5%, HR 72 bpm, suit pressure nominal."
}

# LLM provider: "ollama" (default), "afm", "llamacpp"
LLM_PROVIDER=ollama
LLM_MODEL=gemma4:26b
LLM_API_BASE=              # auto-set per provider if empty

# Ollama host -- always localhost. If OLLAMA_HOST is set system-wide to 0.0.0.0
# (Ollama's server bind address), the orchestrator automatically remaps it to localhost.
OLLAMA_HOST=http://localhost:11434
OLLAMA_KEEP_ALIVE=30m
OLLAMA_PREWARM=true

# Tokenizer -- auto-resolved to local tokenizer_cache/ dir if pre-downloaded.
# Startup expects local tokenizer files for offline field use.
# TOKENIZER_NAME=google/gemma-4-26B-A4B-it

# Context management
MAX_CONTEXT_TOKENS=128000
SUMMARIZE_THRESHOLD=80000

# TSS2026 server (UDP telemetry -- used by the get_tss_state MCP tool)
# Set TSS_UDP_HOST to the IP of the running TSS instance.
TSS_UDP_HOST=10.206.64.189
TSS_UDP_PORT=14141
TSS_UDP_TIMEOUT=2.0

tokenizer_cache/
  google--gemma-4-26B-A4B-it/    # Gemma 4 tokenizer (used by ollama + llamacpp providers)
    tokenizer.json
    tokenizer_config.json
    chat_template.jinja

uv run python -c "
from transformers import AutoTokenizer
import os
name = 'google/gemma-4-26B-A4B-it'  # change to the HF model ID
save_dir = os.path.join('tokenizer_cache', name.replace('/', '--'))
os.makedirs(save_dir, exist_ok=True)
tok = AutoTokenizer.from_pretrained(name)
tok.save_pretrained(save_dir)
print(f'Saved to {save_dir}')
"

orchestrator/
  main.py                  FastAPI app, /chat agentic loop
  config.py                All configuration, env var loading, MCP server wiring
  llm_provider.py          Provider abstraction: Ollama, AFM/MLX, llama.cpp
  tss_udp_client.py        Async UDP client for TSS2026 telemetry
  mcp_client.py            MCP server lifecycle, tool discovery, tool execution
  context_manager.py       Token counting, budget calculation, truncation
  context_summarizer.py    Auto-summarizes when conversation exceeds 80k tokens
  tokenizer_cache/         Pre-downloaded HF tokenizer files (offline operation)
  docs/                    Mission/headset references searchable by the LLM (text, PDF, images)
  mcp_servers/
    tss_tools_server.py    MCP server: get_tss_state, search_docs, read_doc, inspect_image
  rag_service.py           Conversation history via LanceDB (disabled, preserved for later)
  test_tss_udp.py          Smoke test for UDP connectivity to TSS2026

# Verify connectivity
uv run python test_tss_udp.py

Magic Leap 2 headset / client -> POST /chat
  |
1. Build message list: [system_prompt] + [user messages]
   |-- System prompt tells Luna about Vosk/MLVoice artifacts, "send recording",
   |   Magic Leap Unity scenes, and the headset docs to consult
2. If total tokens > 80k -> summarize older conversation history
3. Send to Gemma 4 (with think=True)
4. Model reasons internally (thinking trace logged, not streamed)
5. Model decides: answer directly OR call tools
   |-- get_tss_state(scope=eva) -> UDP to TSS2026 -> live JSON
   |-- search_docs("oxygen") -> grep over docs/ -> file:line matches
   |-- search_docs("show reference map") -> exact headset voice/runtime docs
   |-- read_doc("procedures.pdf", around_line=42) -> context excerpt
   |-- inspect_image("maps/dust-map.png") -> base64 injected into LLM context
6. Tool results fed back -> model generates final answer
   |-- If images were injected, base64 is stripped after the LLM responds
   |   (images exist in context for exactly one inference call)
7. Return response (content only, no thinking trace)

# Test TSS2026 UDP connectivity (requires running TSS server)
uv run python test_tss_udp.py

# Quick model test
uv run python -c "
import asyncio
from llm_provider import OllamaProvider
async def test():
    p = OllamaProvider(host='http://localhost:11434')
    r = await p.chat(model='gemma4:26b', messages=[{'role':'user','content':'Hello'}])
    print(f'Content: {r.content}')
    print(f'Thinking: {len(r.thinking)} chars' if r.thinking else 'No thinking')
asyncio.run(test())
"