docs: remove stray 400k references

Author: ndycode

CHANGELOG.md

@@ -4,7 +4,7 @@ All notable changes to this project are documented here. Dates use the ISO forma
 
 ## [3.3.0] - 2025-11-19
 ### Added
-- GPT 5.1 Codex Max support: normalization, per-model defaults, and new presets (`gpt-5.1-codex-max`, `gpt-5.1-codex-max-xhigh`) with expanded output window and `xhigh` reasoning.
+- GPT 5.1 Codex Max support: normalization, per-model defaults, and new presets (`gpt-5.1-codex-max`, `gpt-5.1-codex-max-xhigh`) with extended reasoning options (including `none`/`xhigh`) while keeping the 272k context / 128k output limits.
 - Typing and config support for new reasoning options (`none`/`xhigh`, summary `off`/`on`) plus updated test matrix entries.
 
 ### Changed

README.md

@@ -33,7 +33,7 @@ Follow me on [X @nummanthinks](https://x.com/nummanthinks) for future updates an
 ## Features
 
 - ✅ **ChatGPT Plus/Pro OAuth authentication** - Use your existing subscription
-- ✅ **12 pre-configured GPT 5.1 variants** - GPT 5.1, GPT 5.1 Codex, GPT 5.1 Codex Max, and GPT 5.1 Codex Mini presets for common reasoning levels (including `gpt-5.1-codex-max-low/medium/high/xhigh`)
+- ✅ **13 pre-configured GPT 5.1 variants** - GPT 5.1, GPT 5.1 Codex, GPT 5.1 Codex Max, and GPT 5.1 Codex Mini presets for common reasoning levels (including `gpt-5.1-codex-max` and `gpt-5.1-codex-max-low/medium/high/xhigh`)
 - ⚠️ **GPT 5.1 only** - Older GPT 5.0 models are deprecated and may not work reliably
 - ✅ **Zero external dependencies** - Lightweight with only @openauthjs/openauth
 - ✅ **Auto-refreshing tokens** - Handles token expiration automatically
@@ -134,7 +134,7 @@ Follow me on [X @nummanthinks](https://x.com/nummanthinks) for future updates an
           "name": "GPT 5.1 Codex Max (OAuth)",
           "limit": {
             "context": 272000,
-            "output": 400000
+            "output": 128000
           },
           "options": {
             "reasoningEffort": "high",
@@ -198,7 +198,7 @@ Follow me on [X @nummanthinks](https://x.com/nummanthinks) for future updates an
           "name": "GPT 5.1 Codex Max Extra High (OAuth)",
           "limit": {
             "context": 272000,
-            "output": 400000
+            "output": 128000
           },
           "options": {
             "reasoningEffort": "xhigh",
@@ -299,7 +299,7 @@ Follow me on [X @nummanthinks](https://x.com/nummanthinks) for future updates an
    **Global config**: `~/.config/opencode/opencode.json`
    **Project config**: `<project>/.opencode.json`
 
-   This gives you 12 GPT 5.1 variants with different reasoning levels:
+   This gives you 13 GPT 5.1 variants with different reasoning levels:
    - **gpt-5.1-codex** (low/medium/high) - Latest Codex model presets
    - **gpt-5.1-codex-max** (low/medium/high/xhigh) - Codex Max presets (`gpt-5.1-codex-max-low/medium/high/xhigh`)
    - **gpt-5.1-codex-mini** (medium/high) - Latest Codex mini tier presets
@@ -395,6 +395,7 @@ When using [`config/full-opencode.json`](./config/full-opencode.json), you get t
 | `gpt-5.1-codex-low` | GPT 5.1 Codex Low (OAuth) | Low | Fast code generation |
 | `gpt-5.1-codex-medium` | GPT 5.1 Codex Medium (OAuth) | Medium | Balanced code tasks |
 | `gpt-5.1-codex-high` | GPT 5.1 Codex High (OAuth) | High | Complex code & tools |
+| `gpt-5.1-codex-max` | GPT 5.1 Codex Max (OAuth) | High | Default Codex Max preset with large-context support |
 | `gpt-5.1-codex-max-low` | GPT 5.1 Codex Max Low (OAuth) | Low | Fast exploratory large-context work |
 | `gpt-5.1-codex-max-medium` | GPT 5.1 Codex Max Medium (OAuth) | Medium | Balanced large-context builds |
 | `gpt-5.1-codex-max-high` | GPT 5.1 Codex Max High (OAuth) | High | Long-horizon builds, large refactors |
@@ -410,7 +411,7 @@ When using [`config/full-opencode.json`](./config/full-opencode.json), you get t
 
 > **Note**: All `gpt-5.1-codex-mini*` presets map directly to the `gpt-5.1-codex-mini` slug with standard Codex limits (272k context / 128k output).
 >
-> **Note**: Codex Max presets use the `gpt-5.1-codex-max` slug with 272k input and expanded ~400k output support. Use `gpt-5.1-codex-max-low/medium/high/xhigh` to pick reasoning level (only `-xhigh` uses `xhigh` reasoning).
+> **Note**: Codex Max presets use the `gpt-5.1-codex-max` slug with 272k context and 128k output. Use `gpt-5.1-codex-max-low/medium/high/xhigh` to pick reasoning level (only `-xhigh` uses `xhigh` reasoning).
 
 > **⚠️ Important**: GPT 5 models can be temperamental - some variants may work better than others, some may give errors, and behavior may vary. Stick to the presets above configured in `full-opencode.json` for best results.
 
@@ -482,7 +483,7 @@ If you want to customize settings yourself, you can configure options at provide
 > **Notes**:
 > - `minimal` effort is auto-normalized to `low` for Codex models.
 > - Codex Mini clamps to `medium`/`high`; `xhigh` downgrades to `high`.
-> - Codex Max supports `none`/`xhigh` plus expanded output limits (~400k).
+> - Codex Max supports `none`/`xhigh` plus extended reasoning options while keeping the same 272k context / 128k output limits.
 
 #### Global Configuration Example
 

config/README.md

@@ -22,7 +22,7 @@ cp config/full-opencode.json ~/.config/opencode/opencode.json
 **What's included:**
 - All supported GPT 5.1 variants: gpt-5.1, gpt-5.1-codex, gpt-5.1-codex-max, gpt-5.1-codex-mini
 - Proper reasoning effort settings for each variant (including new `xhigh` for Codex Max)
-- Context limits (272k context / 128k output for core Codex; Codex Max allows larger outputs)
+- Context limits (272k context / 128k output for all Codex families, including Codex Max)
 - Required options: `store: false`, `include: ["reasoning.encrypted_content"]`
 
 ### ❌ Other Configurations (NOT SUPPORTED)

docs/configuration.md

@@ -394,7 +394,7 @@ CODEX_MODE=1 opencode run "task"  # Temporarily enable
 ## Configuration Files
 
 **Provided Examples:**
-- [config/full-opencode.json](../config/full-opencode.json) - Complete with 8 GPT 5.1 variants
+- [config/full-opencode.json](../config/full-opencode.json) - Complete with 13 GPT 5.1 variants
 
 > **⚠️ REQUIRED:** You MUST use `full-opencode.json` - this is the ONLY officially supported configuration. Minimal configs are NOT supported for GPT 5 models and will fail unpredictably. OpenCode's auto-compaction and usage widgets also require the full config's per-model `limit` metadata.
 

docs/getting-started.md

@@ -98,7 +98,7 @@ Add this to `~/.config/opencode/opencode.json`:
           "name": "GPT 5.1 Codex Max (OAuth)",
           "limit": {
             "context": 272000,
-            "output": 400000
+            "output": 128000
           },
           "options": {
             "reasoningEffort": "high",
@@ -154,7 +154,7 @@ Add this to `~/.config/opencode/opencode.json`:
           "name": "GPT 5.1 Codex Max Extra High (OAuth)",
           "limit": {
             "context": 272000,
-            "output": 400000
+            "output": 128000
           },
           "options": {
             "reasoningEffort": "xhigh",
@@ -242,16 +242,16 @@ Add this to `~/.config/opencode/opencode.json`:
 
   **What you get:**
   - ✅ GPT 5.1 Codex (Low/Medium/High reasoning)
-  - ✅ GPT 5.1 Codex Max (Low/Medium/High/xHigh reasoning presets, larger outputs)
+  - ✅ GPT 5.1 Codex Max (Low/Medium/High/xHigh reasoning presets)
   - ✅ GPT 5.1 Codex Mini (Medium/High reasoning)
   - ✅ GPT 5.1 (Low/Medium/High reasoning)
-  - ✅ 272k context + 128k output window for core presets (Codex Max expands output to ~400k)
+  - ✅ 272k context + 128k output window for all GPT 5.1 presets.
   - ✅ All visible in OpenCode model selector
   - ✅ Optimal settings for each reasoning level
 
 > **Note**: All `gpt-5.1-codex-mini*` presets use 272k context / 128k output limits.
 >
-> **Note**: Codex Max presets map to the `gpt-5.1-codex-max` slug with 272k input and expanded ~400k output. Use `gpt-5.1-codex-max-low/medium/high/xhigh` to pick the reasoning level (only `-xhigh` uses `xhigh` reasoning).
+> **Note**: Codex Max presets map to the `gpt-5.1-codex-max` slug with 272k context and 128k output. Use `gpt-5.1-codex-max-low/medium/high/xhigh` to pick the reasoning level (only `-xhigh` uses `xhigh` reasoning).
 
 Prompt caching is enabled out of the box: when OpenCode sends its session identifier as `prompt_cache_key`, the plugin forwards it untouched so multi-turn runs reuse prior work. The CODEX_MODE bridge prompt bundled with the plugin is kept in sync with the latest Codex CLI release, so the OpenCode UI and Codex share the same tool contract. If you hit your ChatGPT subscription limits, the plugin returns a friendly Codex-style message with the 5-hour and weekly usage windows so you know when capacity resets.
 
@@ -299,7 +299,7 @@ opencode run "write hello world to test.txt" --model=openai/gpt-5.1-codex-medium
 opencode
 ```
 
-You'll see all 8 GPT 5.1 variants (5.1, 5.1 Codex, and 5.1 Codex Mini presets) in the model selector!
+You'll see all 13 GPT 5.1 variants (Codex, Codex Max, Codex Mini, and GPT 5.1 presets) in the model selector!
 
 ---