Update models docs

Author: jahooma

web/src/content/advanced/what-models.mdx

@@ -18,7 +18,7 @@ The main agent ("Buffy") coordinates everything:
 |------|-------|
 | Default | Claude Opus 4.5 |
 | Max | Claude Opus 4.5 |
-| Lite | Grok 4.1 Fast |
+| Free | Grok 4.1 Fast |
 </MarkdownTable>
 
 ## Subagents
@@ -28,15 +28,14 @@ The orchestrator spawns these for specific jobs:
 <MarkdownTable>
 | Task | Models |
 |------|--------|
-| Code editing | Claude Opus 4.5, GPT-5.1 |
-| Thinking/reasoning | Claude Opus 4.5, GPT-5.1, Gemini 2.5 Pro |
-| Code review | Claude Opus 4.5, Claude Sonnet 4.5, GPT-5.1 |
-| File discovery | Gemini 2.0 Flash, Grok 4 Fast |
-| Terminal commands | Grok 4 Fast, Claude Sonnet 4.5 |
-| Context management | GPT-5 Mini |
-| Web/docs research | Grok 4 Fast |
+| Code editing | Claude Opus 4.5, GLM 4.7 |
+| Thinking/reasoning | Claude Opus 4.5, GPT-5.2 |
+| Code review | Claude Opus 4.5, GPT-5.2 |
+| File discovery | Gemini 2.5 Flash Lite, Grok 4.1 Fast |
+| Terminal commands | Claude Haiku 4.5, Grok 4.1 Fast |
+| Web/docs research | Grok 4.1 Fast |
 </MarkdownTable>
 
-Max mode runs multiple implementations in parallel and picks the best one. Default mode runs a single implementation pass. Lite mode skips validation steps for speed.
+Max mode runs multiple implementations in parallel and picks the best one. Default mode runs a single implementation pass. Free mode uses GLM 4.7 for editing.
 
 File rewrites use speculative decoding from Relace AI.