haylan/LLM-Server
1
0 Fork 0
Code Issues 4 Pull Requests Actions Packages Projects Releases Wiki Activity

feat(llm): add --spec-draft-p-min 0.5 to DFlash2 spec args

Browse Source 
DFlash2 truncates its draft when selector confidence drops below p_min,
cutting wasted verify work on long, low-acceptance thinking output.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
haylanandClaude Bot committed 2026-11-06 18:53:27 +01:00
1 parent 6ce527741b
commit 51684fb7a4
2 files changed
+2 -2
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

.env.example
+1 -1
View File
@@ -9,7 +9,7 @@ LLAMA_CACHE_RAM=0
LLAMA_KV_CACHE_TYPE=q8_0
# Blank = speculative decoding off; see docs/architecture.md#llama-server
LLAMA_SPEC_ARGS=--spec-type draft-dflash -md /models/Qwen3.8-27B-DFlash2-Q8_0.gguf --spec-draft-n-max 3
LLAMA_SPEC_ARGS=--spec-type draft-dflash -md /models/Qwen3.8-27B-DFlash2-Q8_0.gguf --spec-draft-n-max 3 --spec-draft-p-min 0.5
# --- SearXNG web search --- see docs/gateway.md#web-search
# Auto-filled by update.sh — leave blank.