zeroclaw, variablize the context amount to be per-machine.

tasks/workstation/linux/software/zeroclaw.yml

@@ -43,6 +43,8 @@
         # Conversely:
         #   - qwen3.5:4b would be 3.3GB * 1.3 = 4290, rather than 4000.
         llm_size_coef: 1.3
+        # Take the GB of RAM and use it as the MB of context.
+        llm_context: "{{ ansible_memtotal_mb | int }}"
 
     - name: Workstation | Linux | Software | ZeroClaw | Facts 02
       ansible.builtin.set_fact:
@@ -181,7 +183,7 @@
               export OLLAMA_KEEP_ALIVE=20m # default 5m
               export OLLAMA_MAX_LOADED_MODELS={{ llm_concurrency }} # default is 3 if using CPUs, or 3 * NumGPUs
               export OLLAMA_NUM_PARALLEL={{ llm_parallel }} # default is 1 and applies per-model (unconfirmed)
-              export OLLAMA_CONTEXT_LENGTH=32000 # default is 4096, https://docs.ollama.com/faq#how-can-i-specify-the-context-window-size
+              export OLLAMA_CONTEXT_LENGTH={{ llm_context }} # default is 4096, https://docs.ollama.com/faq#how-can-i-specify-the-context-window-size
               export OLLAMA_MAX_QUEUE=999 # default is 512 (unconfirmed)
               exec chpst -u {{ user }} ollama serve
             dest: "{{ ollama_runit_script }}"
@@ -364,7 +366,7 @@
           [providers.models.ollama.{{ llm.provider }}]
           model = "{{ llm.model }}"
           timeout_secs = 3600 # 1 hour, default is 120 (2 minutes), issue happens between 5 and 6 minutes though.
-          context_window = 999999999
+          context_window = {{ llm_context }}
           {% endfor %}
         state: present
         create: false