{% extends "base.html" %} {% block title %}Models — Agent Wasp{% endblock %} {% block page_css %} {% endblock %} {% block content %} {# ── Provider brand registry ──────────────────────────────────────────────── #} {% set PROV = { "openai": {"color":"#10a37f","label":"OAI","desc":"GPT-5.4 · o3 · GPT-4o", "icon":"dark/openai"}, "anthropic": {"color":"#D4581A","label":"ANT","desc":"Claude 4 · Claude 3.5", "icon":"light/claude-color"}, "google": {"color":"#4285F4","label":"GGL","desc":"Gemini 3 · 2.5 · 2.0", "icon":"light/gemini-color"}, "xai": {"color":"#BBBBBB","label":"xAI","desc":"Grok 4 · Grok 2", "icon":"dark/grok"}, "mistral": {"color":"#FF7000","label":"MST","desc":"Mistral · Magistral · Code", "icon":"light/mistral-color"}, "deepseek": {"color":"#5B6CF6","label":"DSK","desc":"DeepSeek R1 · V3", "icon":"light/deepseek-color"}, "openrouter": {"color":"#9B59B6","label":"OR", "desc":"100+ models", "icon":"dark/openrouter"}, "perplexity": {"color":"#20B2AA","label":"PPX","desc":"Sonar Pro · Reasoning", "icon":"light/perplexity-color"}, "huggingface": {"color":"#FFD21E","label":"HF", "desc":"Open source models", "icon":"light/huggingface-color"}, "lmstudio": {"color":"#8B5CF6","label":"LMS","desc":"Local (LM Studio)", "icon":""}, "ollama": {"color":"#6EE7B7","label":"OLM","desc":"Local inference", "icon":"dark/ollama"}, "moonshot": {"color":"#6366F1","label":"KIM","desc":"Kimi K2 · Moonshot", "icon":"dark/moonshot"}, } %} {# ── Lobe icons base ──────────────────────────────────────────────────────── #} {% set ICON_BASE = "https://raw.githubusercontent.com/lobehub/lobe-icons/refs/heads/master/packages/static-png/" %} {# ── Active provider meta ─────────────────────────────────────────────────── #} {% set ap = PROV.get(status.active_provider, {"color":"#6B7280","label":"?","desc":"Unknown","icon":""}) %}
{{ installed|length }} models · {{ status.providers|length }} active provider{{ 's' if status.providers|length != 1 }}
| Model | Size | Status | Active | Default | Uninstall | ||
|---|---|---|---|---|---|---|---|
|
{% if mp.icon %}
{{ m.name }}
{% if m.is_default %}★{% endif %}
|
{% if mp.icon %}
|
{% if m.size %} {{ m.size }} {% else %} cloud {% endif %} | {% if m.active %} Active {% else %} Idle {% endif %} | {% if m.provider == "ollama" %} | {% else %} | {% endif %} |
Ollama service is not running
Local models let the agent run inference on this machine without sending data to a cloud API. Disk usage: ~4–8 GB per model; RAM: ≥8 GB recommended.
To enable, run on the host:
echo 'COMPOSE_PROFILES=local-llm' >> /opt/wasp/.env && cd /opt/wasp && docker compose up -d agent-ollama
After Ollama starts, refresh this page — the "Install Models" button will appear.
| Model | Size | Status | Active | Uninstall |
|---|---|---|---|---|
| {{ m.name }} | {% if m.size %} {{ m.size }} {% else %} — {% endif %} | {% if m.active %} Active {% else %} Idle {% endif %} |
No models installed yet. Use the catalog below to download one.
{% endif %}