koko210Serve
eafab336b4
docker-compose.yml: - Add Traefik proxy network + labels for miku.panel domain - Connect miku-bot service to proxy network llama-swap-config.yaml / llama-swap-rocm-config.yaml: - Add --chat-template-file flag to disable Llama 3.1 built-in tool calling (was causing malformed responses) - ROCm config: add Rocinante-X 12B model entry for comparison testing cheshire-cat discord_bridge plugin: - Increase declarative memory recall (k=3→10, threshold 0.7→0.5) for better factual retrieval - Add agent_prompt_prefix hook to enforce factual accuracy from declarative memories - Add before_agent_starts debug logging for memory inspection - Add passthrough hooks for message/suffix pipeline
44 lines
1.9 KiB
YAML
44 lines
1.9 KiB
YAML
# llama-swap configuration for AMD RX 6800 (ROCm)
|
|
# This manages automatic model switching and unloading for the secondary GPU
|
|
# Vision model stays on NVIDIA GPU - AMD only handles text models
|
|
|
|
models:
|
|
# Main text generation model (same name as NVIDIA for uniform switching)
|
|
# Custom chat template to disable built-in tool calling
|
|
llama3.1:
|
|
cmd: /app/llama-server --port ${PORT} --model /models/Llama-3.1-8B-Instruct-UD-Q4_K_XL.gguf -ngl 99 -c 16384 --host 0.0.0.0 --no-warmup --flash-attn on --chat-template-file /app/llama31_notool_template.jinja
|
|
ttl: 1800 # Unload after 30 minutes of inactivity (1800 seconds)
|
|
aliases:
|
|
- llama3.1
|
|
- text-model
|
|
|
|
# Evil/Uncensored model (same name as NVIDIA for uniform switching)
|
|
darkidol:
|
|
cmd: /app/llama-server --port ${PORT} --model /models/DarkIdol-Llama-3.1-8B-Instruct-1.3-Uncensored_Q4_K_M.gguf -ngl 99 -c 16384 --host 0.0.0.0 --no-warmup --flash-attn on
|
|
ttl: 1800 # Unload after 30 minutes of inactivity
|
|
aliases:
|
|
- darkidol
|
|
- evil-model
|
|
- uncensored
|
|
|
|
# Rocinante-X 12B - larger creative/RP model for comparison testing
|
|
rocinante:
|
|
cmd: /app/llama-server --port ${PORT} --model /models/Rocinante-X-12B-v1b-Q5_K_M.gguf -ngl 99 -c 8192 --host 0.0.0.0 --no-warmup --flash-attn on
|
|
ttl: 1800 # Unload after 30 minutes of inactivity
|
|
aliases:
|
|
- rocinante
|
|
- rocinante-12b
|
|
|
|
# Japanese language model (Llama 3.1 Swallow - Japanese optimized)
|
|
swallow:
|
|
cmd: /app/llama-server --port ${PORT} --model /models/Llama-3.1-Swallow-8B-Instruct-v0.5-Q4_K_M.gguf -ngl 99 -c 16384 --host 0.0.0.0 --no-warmup --flash-attn on
|
|
ttl: 1800 # Unload after 30 minutes of inactivity
|
|
aliases:
|
|
- swallow
|
|
- japanese
|
|
- japanese-model
|
|
|
|
# Server configuration
|
|
# llama-swap will listen on this address
|
|
# Inside Docker, we bind to 0.0.0.0 to allow bot container to connect
|