Experimental/brain-of-reese
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(rag): lift chat output cap to 32768 tokens — long answers no longer cut off

Browse Source
This commit is contained in:
ducoterra
2026-08-22 11:30:19 -04:00
parent 6ec6181c7b
commit 0da5275eeb
9 changed files with 295 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.env.example
+1
View File
@@ -21,6 +21,7 @@ BOR_EMBEDDING_DIM=768 # verified 2026-08 via scripts/llm_probe.py
BOR_TOP_N_DOCS=2
BOR_RELEVANCE_THRESHOLD=0.62 # answer when best cosine >= this OR an FTS hit; else honest deflection
BOR_MAX_CONTEXT_CHARS=24000 # cap on total document text sent to the LLM
BOR_MAX_OUTPUT_TOKENS=32768 # max answer length in tokens (answers must not be cut off)
BOR_CHUNK_TARGET_CHARS=2000
BOR_CHUNK_OVERLAP_CHARS=200
BOR_EMBED_BATCH_SIZE=16