agent: repoint FLM-NPU model roles off the retired provider; sync memory/skills
config.yml: vision -> llamacpp (dankclaw)/frank-large:auto, tiny -> llamacpp (dankserver)/frank-small:off, task -> llamacpp (dankclaw)/frank-large:medium (was flm dankclaw-npu), matching the flm (dankclaw-npu) provider removal in models.yml. Memories/managed skills: rollouts and learned notes from the frank-large KV-pool + sampler work and the noctalia recolor; merged the duplicate noctalia-palette-recolor skill into noctalia-custom-palette-recolor (adds the dotfiles-tracking step and the +show-config delta gotcha) and dropped the superseded legacy memory skills whose ghostty USR1 reload recipe is wrong for this fleet.
This commit is contained in:
1 parent
c32584e093
commit
872e86399a
56 files changed
+318
-178
No files matched your search
@@ -0,0 +1 @@
|
||||
Manages frank-large/small inference routing, FLM NPU constraints, and Halogen runtime kwargs for thinking block preservation.
|
||||
+1
@@ -0,0 +1 @@
|
||||
Pinned sampler: temp=1.0, top_p=0.95, top_k=20. Live config read on next cycle.
|
||||
+1
@@ -0,0 +1 @@
|
||||
ls -l ~/.config/flm/xclbins/ | grep -E '\.xclbin' || echo 'Missing symlinks'
|
||||
+1
@@ -0,0 +1 @@
|
||||
curl -s localhost:7771/status | jq .queue
|
||||
+1
@@ -0,0 +1 @@
|
||||
enable_thinking: true\npreserve_thinking: true\nreasoning_effort: medium
|
||||
+1
@@ -0,0 +1 @@
|
||||
{"thinkingLevelMap": true, "gemma4-it": "e4b"}
|
||||
Reference in new issue
Block a user