fix: limite de CPU 6.0 para ollama

803% de CPU durante el thinking (ya fixeado con think=false) podria
repetirse en cargas largas. Cap de 6 CPUs (de las 8+ del host):
- compose (openwebui/wordpress): deploy.resources.limits.cpus: 6.0
- docker run (hermes/openclaw): --cpus 6.0

Verificado: cap de 3.5 en host de 4 cores => 353% max (cap respetado).

hermes/deploy.ps1

@@ -117,7 +117,7 @@ $running = docker ps --format '{{.Names}}' | Select-String -Quiet 'd0a1-hermes-o
 if (-not $running) {
     $GpuArgs = @()
     if ($HasGpu) { $GpuArgs = @('--gpus', 'all') }
-    docker run -d --name d0a1-hermes-ollama --memory 6g @GpuArgs `
+    docker run -d --name d0a1-hermes-ollama --memory 6g --cpus 6.0 @GpuArgs `
         -p "${OllamaPort}:11434" `
         -v d0a1_hermes_ollama:/root/.ollama `
         -e OLLAMA_HOST=0.0.0.0 `
@@ -147,6 +147,7 @@ if ($env:HTTPS_PROXY) { $ProxyEnv = @('-e', "HTTPS_PROXY=$env:HTTPS_PROXY", '-e'
 $ModelList = docker exec d0a1-hermes-ollama ollama list 2>$null
 if ("$ModelList" -match 'qwen3:4b') {
     Info 'Modelo qwen3:4b ya descargado (volumen intacto). Omitiendo descarga.'
+    cmd /c docker exec d0a1-hermes-ollama sh -c 'printf "FROM qwen3:4b\\nPARAMETER num_ctx 16384\\nSYSTEM /no_think\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf' '>' NUL '2>&1'
 } else {
     # --- Modelo: HF primero (CDN propio, rapido) y registry de Ollama como ultimo recurso ---
 
@@ -165,7 +166,7 @@ foreach ($Url in $HfUrls) {
 if ((Test-Path $Gguf) -and (Get-Item $Gguf).Length -gt 2400000000) {
     Info 'Importando modelo en Ollama...'
     docker cp "$Gguf" d0a1-hermes-ollama:/root/.ollama/import.gguf
-    docker exec d0a1-hermes-ollama sh -c 'printf "FROM /root/.ollama/import.gguf\\nPARAMETER num_ctx 16384\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
+    docker exec d0a1-hermes-ollama sh -c 'printf "FROM /root/.ollama/import.gguf\\nPARAMETER num_ctx 16384\\nSYSTEM /no_think\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
     $CreateOk = ($LASTEXITCODE -eq 0)
     docker exec d0a1-hermes-ollama rm -f /root/.ollama/import.gguf | Out-Null
     Remove-Item "$Gguf" -Force -ErrorAction SilentlyContinue
@@ -178,7 +179,7 @@ if ((Test-Path $Gguf) -and (Get-Item $Gguf).Length -gt 2400000000) {
     if ($LASTEXITCODE -eq 0) {
         $PullOk = $true
         # Contexto ampliado (el default 4096 da 400 en Open WebUI con prompts largos)
-        docker exec d0a1-hermes-ollama sh -c 'printf "FROM qwen3:4b\\nPARAMETER num_ctx 16384\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
+        docker exec d0a1-hermes-ollama sh -c 'printf "FROM qwen3:4b\\nPARAMETER num_ctx 16384\\nSYSTEM /no_think\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
     }
 }
 if (-not $PullOk) {
@@ -191,6 +192,10 @@ if (-not $PullOk) {
     exit 1
 }
 }
+# --- Precalentar el modelo (load ya, no en el primer mensaje) ---------------
+Warn '        Precargando el modelo en memoria...'
+cmd /c docker exec d0a1-hermes-ollama sh -c 'ollama run qwen3:4b ""' '>' NUL '2>&1'
+
 Ok "Ollama corriendo en http://localhost:${OllamaPort}"
 
 # --- 2. Instalar Hermes Agent ------------------------------------------------

hermes/remove.ps1

@@ -0,0 +1,16 @@
+# ============================================================
+#   d0a1 pack - Hermes (Windows) - DESINSTALAR
+#   Elimina el contenedor y el volumen (datos + modelo).
+# ============================================================
+
+$ErrorActionPreference = 'Stop'
+
+function Info($m) { Write-Host ">>> $m" }
+
+Info 'Borrando contenedor y volumen...'
+cmd /c docker rm -f d0a1-hermes-ollama '>' NUL '2>&1'
+cmd /c docker volume rm d0a1_hermes_ollama '>' NUL '2>&1'
+
+Write-Host ''
+Info 'Eliminado TODO: contenedor + volumen (incluido el modelo IA).'
+Write-Host ''

hermes/remove.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+# d0a1 pack - Hermes (Linux) - DESINSTALAR
+docker rm -f d0a1-hermes-ollama >/dev/null 2>&1 || true
+docker volume rm d0a1_hermes_ollama >/dev/null 2>&1 || true
+echo ">>> Eliminado TODO: contenedor + volumen (incluido el modelo IA)."

openclaw/deploy.ps1

@@ -117,7 +117,7 @@ $running = docker ps --format '{{.Names}}' | Select-String -Quiet 'd0a1-openclaw
 if (-not $running) {
     $GpuArgs = @()
     if ($HasGpu) { $GpuArgs = @('--gpus', 'all') }
-    docker run -d --name d0a1-openclaw-ollama --memory 6g @GpuArgs `
+    docker run -d --name d0a1-openclaw-ollama --memory 6g --cpus 6.0 @GpuArgs `
         -p "${OllamaPort}:11434" `
         -v d0a1_openclaw_ollama:/root/.ollama `
         -e OLLAMA_HOST=0.0.0.0 `
@@ -147,6 +147,7 @@ if ($env:HTTPS_PROXY) { $ProxyEnv = @('-e', "HTTPS_PROXY=$env:HTTPS_PROXY", '-e'
 $ModelList = docker exec d0a1-openclaw-ollama ollama list 2>$null
 if ("$ModelList" -match 'qwen3:4b') {
     Info 'Modelo qwen3:4b ya descargado (volumen intacto). Omitiendo descarga.'
+    cmd /c docker exec d0a1-openclaw-ollama sh -c 'printf "FROM qwen3:4b\\nPARAMETER num_ctx 16384\\nSYSTEM /no_think\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf' '>' NUL '2>&1'
 } else {
     # --- Modelo: HF primero (CDN propio, rapido) y registry de Ollama como ultimo recurso ---
 
@@ -165,7 +166,7 @@ foreach ($Url in $HfUrls) {
 if ((Test-Path $Gguf) -and (Get-Item $Gguf).Length -gt 2400000000) {
     Info 'Importando modelo en Ollama...'
     docker cp "$Gguf" d0a1-openclaw-ollama:/root/.ollama/import.gguf
-    docker exec d0a1-openclaw-ollama sh -c 'printf "FROM /root/.ollama/import.gguf\\nPARAMETER num_ctx 16384\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
+    docker exec d0a1-openclaw-ollama sh -c 'printf "FROM /root/.ollama/import.gguf\\nPARAMETER num_ctx 16384\\nSYSTEM /no_think\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
     $CreateOk = ($LASTEXITCODE -eq 0)
     docker exec d0a1-openclaw-ollama rm -f /root/.ollama/import.gguf | Out-Null
     Remove-Item "$Gguf" -Force -ErrorAction SilentlyContinue
@@ -178,7 +179,7 @@ if ((Test-Path $Gguf) -and (Get-Item $Gguf).Length -gt 2400000000) {
     if ($LASTEXITCODE -eq 0) {
         $PullOk = $true
         # Contexto ampliado (el default 4096 da 400 en Open WebUI con prompts largos)
-        docker exec d0a1-openclaw-ollama sh -c 'printf "FROM qwen3:4b\\nPARAMETER num_ctx 16384\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
+        docker exec d0a1-openclaw-ollama sh -c 'printf "FROM qwen3:4b\\nPARAMETER num_ctx 16384\\nSYSTEM /no_think\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
     }
 }
 if (-not $PullOk) {
@@ -191,6 +192,10 @@ if (-not $PullOk) {
     exit 1
 }
 }
+# --- Precalentar el modelo (load ya, no en el primer mensaje) ---------------
+Warn '        Precargando el modelo en memoria...'
+cmd /c docker exec d0a1-openclaw-ollama sh -c 'ollama run qwen3:4b ""' '>' NUL '2>&1'
+
 Ok "Ollama corriendo en http://localhost:${OllamaPort}"
 
 # --- 2. Instalar OpenClaw -----------------------------------------------------

openclaw/remove.ps1

@@ -0,0 +1,16 @@
+# ============================================================
+#   d0a1 pack - OpenClaw (Windows) - DESINSTALAR
+#   Elimina el contenedor y el volumen (datos + modelo).
+# ============================================================
+
+$ErrorActionPreference = 'Stop'
+
+function Info($m) { Write-Host ">>> $m" }
+
+Info 'Borrando contenedor y volumen...'
+cmd /c docker rm -f d0a1-openclaw-ollama '>' NUL '2>&1'
+cmd /c docker volume rm d0a1_openclaw_ollama '>' NUL '2>&1'
+
+Write-Host ''
+Info 'Eliminado TODO: contenedor + volumen (incluido el modelo IA).'
+Write-Host ''

openclaw/remove.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+# d0a1 pack - OpenClaw (Linux) - DESINSTALAR
+docker rm -f d0a1-openclaw-ollama >/dev/null 2>&1 || true
+docker volume rm d0a1_openclaw_ollama >/dev/null 2>&1 || true
+echo ">>> Eliminado TODO: contenedor + volumen (incluido el modelo IA)."

openwebui/deploy.ps1

@@ -168,6 +168,7 @@ if ($env:HTTPS_PROXY) { $ProxyEnv = @('-e', "HTTPS_PROXY=$env:HTTPS_PROXY", '-e'
 $ModelList = docker exec d0a1-ollama ollama list 2>$null
 if ("$ModelList" -match 'qwen3:4b') {
     Info 'Modelo qwen3:4b ya descargado (volumen intacto). Omitiendo descarga.'
+    cmd /c docker exec d0a1-ollama sh -c 'printf "FROM qwen3:4b\\nPARAMETER num_ctx 16384\\nSYSTEM /no_think\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf' '>' NUL '2>&1'
 } else {
     # --- Modelo: HF primero (CDN propio, rapido) y registry de Ollama como ultimo recurso ---
 
@@ -186,7 +187,7 @@ foreach ($Url in $HfUrls) {
 if ((Test-Path $Gguf) -and (Get-Item $Gguf).Length -gt 2400000000) {
     Info 'Importando modelo en Ollama...'
     docker cp "$Gguf" d0a1-ollama:/root/.ollama/import.gguf
-    docker exec d0a1-ollama sh -c 'printf "FROM /root/.ollama/import.gguf\\nPARAMETER num_ctx 16384\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
+    docker exec d0a1-ollama sh -c 'printf "FROM /root/.ollama/import.gguf\\nPARAMETER num_ctx 16384\\nSYSTEM /no_think\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
     $CreateOk = ($LASTEXITCODE -eq 0)
     docker exec d0a1-ollama rm -f /root/.ollama/import.gguf | Out-Null
     Remove-Item "$Gguf" -Force -ErrorAction SilentlyContinue
@@ -199,7 +200,7 @@ if ((Test-Path $Gguf) -and (Get-Item $Gguf).Length -gt 2400000000) {
     if ($LASTEXITCODE -eq 0) {
         $PullOk = $true
         # Contexto ampliado (el default 4096 da 400 en Open WebUI con prompts largos)
-        docker exec d0a1-ollama sh -c 'printf "FROM qwen3:4b\\nPARAMETER num_ctx 16384\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
+        docker exec d0a1-ollama sh -c 'printf "FROM qwen3:4b\\nPARAMETER num_ctx 16384\\nSYSTEM /no_think\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
     }
 }
 if (-not $PullOk) {
@@ -213,6 +214,10 @@ if (-not $PullOk) {
 }
 }
 
+# --- Precalentar el modelo (load a GPU/RAM ya, no en el primer mensaje) -----
+Info 'Precargando el modelo en memoria...'
+cmd /c docker exec d0a1-ollama sh -c 'ollama run qwen3:4b ""' '>' NUL '2>&1'
+
 # --- Done -------------------------------------------------------------------
 Write-Host ''
 Info 'Listo! Tu IA esta corriendo.'

openwebui/docker-compose.yml

@@ -18,7 +18,7 @@ services:
       - ollama_data:/root/.ollama
     environment:
       OLLAMA_HOST: "0.0.0.0"
-      OLLAMA_KEEP_ALIVE: "5m"
+      OLLAMA_KEEP_ALIVE: "-1"
       OLLAMA_NUM_PARALLEL: "1"
     healthcheck:
       test: ["CMD", "ollama", "ps"]
@@ -30,6 +30,7 @@ services:
       resources:
         limits:
           memory: 6G
+          cpus: "6.0"
 
   openwebui:
     container_name: d0a1-openwebui
@@ -48,7 +49,7 @@ services:
       ENABLE_WEB_SEARCH: "false"
       ENABLE_DIRECT_CONNECTIONS: "false"
       CHAT_RESPONSE_MAX_TOOL_CALL_ITERATIONS: "1"
-      DEFAULT_MODEL_PARAMS: '{"function_calling": "legacy"}'
+      DEFAULT_MODEL_PARAMS: '{"function_calling": "legacy", "think": false}'
     depends_on:
       ollama:
         condition: service_healthy

openwebui/remove.ps1

@@ -0,0 +1,24 @@
+# ============================================================
+#   d0a1 pack - Open WebUI (Windows) - DESINSTALAR
+#   Elimina contenedores, red y volumenes (datos + modelo).
+# ============================================================
+
+$ErrorActionPreference = 'Stop'
+$ScriptDir = $PSScriptRoot
+if (-not $ScriptDir) { $ScriptDir = Split-Path -Parent $MyInvocation.MyCommand.Path }
+$ComposeFile = Join-Path $ScriptDir 'docker-compose.yml'
+
+function Info($m) { Write-Host ">>> $m" }
+function Err($m)  { Write-Host "!!! $m"; exit 1 }
+
+# Sin confirmacion: este pack no guarda datos sensibles (solo chat + modelo).
+
+Info 'Deteniendo y borrando contenedores...'
+cmd /c docker compose -f "$ComposeFile" down -v --remove-orphans '>' NUL '2>&1'
+if ($LASTEXITCODE -ne 0) { Err 'Fallo el compose down (docker no arrancado?)' }
+
+Write-Host ''
+Info 'Eliminado TODO:'
+Write-Host '  - Contenedores y red'
+Write-Host '  - Volumenes (incluido el modelo IA)'
+Write-Host ''

openwebui/remove.sh

@@ -0,0 +1,9 @@
+#!/usr/bin/env bash
+# d0a1 pack - Open WebUI (Linux) - DESINSTALAR
+set -e
+cd "$(cd "$(dirname "$0")" && pwd)"
+
+echo ">>> Deteniendo y borrando contenedores..."
+docker compose -f docker-compose.yml down -v --remove-orphans >/dev/null 2>&1 || { echo "!!! Fallo el compose down"; exit 1; }
+echo ""
+echo ">>> Eliminado TODO: contenedores, red y volumenes (incluido el modelo IA)."

wordpress/deploy.ps1

@@ -205,6 +205,7 @@ if ($env:HTTPS_PROXY) { $ProxyEnv = @('-e', "HTTPS_PROXY=$env:HTTPS_PROXY", '-e'
 $ModelList = docker exec d0a1-wp-ollama ollama list 2>$null
 if ("$ModelList" -match 'qwen3:4b') {
     Info 'Modelo qwen3:4b ya descargado (volumen intacto). Omitiendo descarga.'
+    cmd /c docker exec d0a1-wp-ollama sh -c 'printf "FROM qwen3:4b\\nPARAMETER num_ctx 16384\\nSYSTEM /no_think\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf' '>' NUL '2>&1'
 } else {
     # --- Modelo: HF primero (CDN propio, rapido) y registry de Ollama como ultimo recurso ---
 
@@ -223,7 +224,7 @@ foreach ($Url in $HfUrls) {
 if ((Test-Path $Gguf) -and (Get-Item $Gguf).Length -gt 2400000000) {
     Info 'Importando modelo en Ollama...'
     docker cp "$Gguf" d0a1-wp-ollama:/root/.ollama/import.gguf
-    docker exec d0a1-wp-ollama sh -c 'printf "FROM /root/.ollama/import.gguf\\nPARAMETER num_ctx 16384\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
+    docker exec d0a1-wp-ollama sh -c 'printf "FROM /root/.ollama/import.gguf\\nPARAMETER num_ctx 16384\\nSYSTEM /no_think\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
     $CreateOk = ($LASTEXITCODE -eq 0)
     docker exec d0a1-wp-ollama rm -f /root/.ollama/import.gguf | Out-Null
     Remove-Item "$Gguf" -Force -ErrorAction SilentlyContinue
@@ -236,7 +237,7 @@ if ((Test-Path $Gguf) -and (Get-Item $Gguf).Length -gt 2400000000) {
     if ($LASTEXITCODE -eq 0) {
         $PullOk = $true
         # Contexto ampliado (el default 4096 da 400 en Open WebUI con prompts largos)
-        docker exec d0a1-wp-ollama sh -c 'printf "FROM qwen3:4b\\nPARAMETER num_ctx 16384\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
+        docker exec d0a1-wp-ollama sh -c 'printf "FROM qwen3:4b\\nPARAMETER num_ctx 16384\\nSYSTEM /no_think\\n" > /tmp/mf && ollama create qwen3:4b -f /tmp/mf'
     }
 }
 if (-not $PullOk) {
@@ -250,6 +251,10 @@ if (-not $PullOk) {
 }
 }
 
+# --- Precalentar el modelo (load a GPU/RAM ya, no en el primer mensaje) -----
+Info 'Precargando el modelo en memoria...'
+cmd /c docker exec d0a1-wp-ollama sh -c 'ollama run qwen3:4b ""' '>' NUL '2>&1'
+
 # --- Done -------------------------------------------------------------------
 Write-Host ''
 Info 'Listo! Tu WordPress con IA esta corriendo.'

wordpress/docker-compose.yml

@@ -52,7 +52,7 @@ services:
       - ollama_data:/root/.ollama
     environment:
       OLLAMA_HOST: "0.0.0.0"
-      OLLAMA_KEEP_ALIVE: "5m"
+      OLLAMA_KEEP_ALIVE: "-1"
     healthcheck:
       test: ["CMD", "ollama", "ps"]
       interval: 15s
@@ -63,6 +63,7 @@ services:
       resources:
         limits:
           memory: 6G
+          cpus: "6.0"
 
   openwebui:
     container_name: d0a1-wp-openwebui
@@ -78,7 +79,7 @@ services:
       WEBUI_AUTH: "false"
       DEFAULT_LOCALE: "es"
       CHAT_RESPONSE_MAX_TOOL_CALL_ITERATIONS: "1"
-      DEFAULT_MODEL_PARAMS: '{"function_calling": "legacy"}'
+      DEFAULT_MODEL_PARAMS: '{"function_calling": "legacy", "think": false}'
     depends_on:
       ollama:
         condition: service_healthy

wordpress/remove.ps1

@@ -0,0 +1,28 @@
+# ============================================================
+#   d0a1 pack - WordPress + IA (Windows) - DESINSTALAR
+#   Elimina contenedores, red y volumenes (datos + modelo).
+# ============================================================
+
+$ErrorActionPreference = 'Stop'
+$ScriptDir = $PSScriptRoot
+if (-not $ScriptDir) { $ScriptDir = Split-Path -Parent $MyInvocation.MyCommand.Path }
+$ComposeFile = Join-Path $ScriptDir 'docker-compose.yml'
+
+function Info($m) { Write-Host ">>> $m" }
+function Err($m)  { Write-Host "!!! $m"; exit 1 }
+
+Write-Host ''
+Write-Host '>>> ATENCION: esto borra TAMBIEN la base de datos y el contenido de WordPress.'
+$Resp = Read-Host 'Continuar? [y/N]'
+if ($Resp -notmatch '^[Yy]') { Write-Host 'Cancelado.'; exit 0 }
+Write-Host ''
+
+Info 'Deteniendo y borrando contenedores...'
+cmd /c docker compose -f "$ComposeFile" down -v --remove-orphans '>' NUL '2>&1'
+if ($LASTEXITCODE -ne 0) { Err 'Fallo el compose down (docker no arrancado?)' }
+
+Write-Host ''
+Info 'Eliminado TODO:'
+Write-Host '  - Contenedores y red'
+Write-Host '  - Volumenes (incluido el modelo IA)'
+Write-Host ''

wordpress/remove.sh

@@ -0,0 +1,12 @@
+#!/usr/bin/env bash
+# d0a1 pack - WordPress + IA (Linux) - DESINSTALAR
+set -e
+cd "$(cd "$(dirname "$0")" && pwd)"
+read -r -p ">>> ATENCION: borra TAMBIEN la BD de WordPress. Continuar? [y/N] " r
+case "$r" in [yY]*) ;; *) echo Cancelado.; exit 0;; esac
+echo ""
+
+echo ">>> Deteniendo y borrando contenedores..."
+docker compose -f docker-compose.yml down -v --remove-orphans >/dev/null 2>&1 || { echo "!!! Fallo el compose down"; exit 1; }
+echo ""
+echo ">>> Eliminado TODO: contenedores, red y volumenes (incluido el modelo IA)."