Merge branch 'dev' of https://github.com/sst/opencode into dev

8 files changed, 92 insertions, 24 deletions

diff --git a/packages/web/src/content/docs/1-0.mdx b/packages/web/src/content/docs/1-0.mdx
index 49db53d72..6737482a7 100644
--- a/packages/web/src/content/docs/1-0.mdx
+++ b/packages/web/src/content/docs/1-0.mdx
@@ -16,13 +16,14 @@ The new TUI works like the old one since it connects to the same opencode server
 You should not be autoupgraded to 1.0 if you are currently using a previous
 version. However some older versions of OpenCode always grab latest.
 
-
 To upgrade manually, run
+
 ```bash
 $ opencode upgrade 1.0.0
 ```
 
 To downgrade back to 0.x, run
+
 ```bash
 $ opencode upgrade 0.15.31
 ```
diff --git a/packages/web/src/content/docs/cli.mdx b/packages/web/src/content/docs/cli.mdx
index 1312442d6..083db369b 100644
--- a/packages/web/src/content/docs/cli.mdx
+++ b/packages/web/src/content/docs/cli.mdx
@@ -184,6 +184,16 @@ This is useful for scripting, automation, or when you want a quick answer withou
 opencode run Explain the use of context in Go
 ```
 
+You can also attach to a running `opencode serve` instance to avoid MCP server cold boot times on every run:
+
+```bash
+# Start a headless server in one terminal
+opencode serve
+
+# In another terminal, run commands that attach to it
+opencode run --attach http://localhost:4096 "Explain async/await in JavaScript"
+```
+
 #### Flags
 
 | Flag         | Short | Description                                                        |
@@ -197,6 +207,8 @@ opencode run Explain the use of context in Go
 | `--file`     | `-f`  | File(s) to attach to message                                       |
 | `--format`   |       | Format: default (formatted) or json (raw JSON events)              |
 | `--title`    |       | Title for the session (uses truncated prompt if no value provided) |
+| `--attach`   |       | Attach to a running opencode server (e.g., http://localhost:4096)  |
+| `--port`     |       | Port for the local server (defaults to random port)                |
 
 ---
 
diff --git a/packages/web/src/content/docs/formatters.mdx b/packages/web/src/content/docs/formatters.mdx
index 8853ca334..9fc41a53d 100644
--- a/packages/web/src/content/docs/formatters.mdx
+++ b/packages/web/src/content/docs/formatters.mdx
@@ -21,9 +21,11 @@ OpenCode comes with several built-in formatters for popular languages and framew
 | clang-format   | .c, .cpp, .h, .hpp, .ino, and [more](https://clang.llvm.org/docs/ClangFormat.html)                       | `.clang-format` config file             |
 | ktlint         | .kt, .kts                                                                                                | `ktlint` command available              |
 | ruff           | .py, .pyi                                                                                                | `ruff` command available with config    |
+| uv             | .py, .pyi                                                                                                | `uv` command available                  |
 | rubocop        | .rb, .rake, .gemspec, .ru                                                                                | `rubocop` command available             |
 | standardrb     | .rb, .rake, .gemspec, .ru                                                                                | `standardrb` command available          |
 | htmlbeautifier | .erb, .html.erb                                                                                          | `htmlbeautifier` command available      |
+| air            | .R                                                                                                       | `air` command available                 |
 
 So if your project has `prettier` in your `package.json`, OpenCode will automatically use it.
 
diff --git a/packages/web/src/content/docs/index.mdx b/packages/web/src/content/docs/index.mdx
index f6d2b3129..20c0d2f99 100644
--- a/packages/web/src/content/docs/index.mdx
+++ b/packages/web/src/content/docs/index.mdx
@@ -41,28 +41,36 @@ You can also install it with the following commands:
 
 - **Using Node.js**
 
-  <Tabs>
+      <Tabs>
+
     <TabItem label="npm">
-      ```bash
-      npm install -g opencode-ai
-      ```
-    </TabItem>
-    <TabItem label="Bun">
+    ```bash 
+    npm install -g opencode-ai 
+    ```
+
+        </TabItem>
+
+      <TabItem label="Bun">
       ```bash
       bun install -g opencode-ai
       ```
-    </TabItem>
-    <TabItem label="pnpm">
+
+        </TabItem>
+
+      <TabItem label="pnpm">
       ```bash
       pnpm install -g opencode-ai
       ```
-    </TabItem>
-    <TabItem label="Yarn">
+
+        </TabItem>
+
+      <TabItem label="Yarn">
       ```bash
       yarn global add opencode-ai
       ```
+
     </TabItem>
-  </Tabs>
+</Tabs>
 
 - **Using Homebrew on macOS and Linux**
 
diff --git a/packages/web/src/content/docs/permissions.mdx b/packages/web/src/content/docs/permissions.mdx
index 4579c2127..a0ba5d5d9 100644
--- a/packages/web/src/content/docs/permissions.mdx
+++ b/packages/web/src/content/docs/permissions.mdx
@@ -187,4 +187,3 @@ permission:
 
 Only analyze code and suggest changes.
 ```
-
diff --git a/packages/web/src/content/docs/providers.mdx b/packages/web/src/content/docs/providers.mdx
index 41e6308f7..6a7520e56 100644
--- a/packages/web/src/content/docs/providers.mdx
+++ b/packages/web/src/content/docs/providers.mdx
@@ -301,6 +301,42 @@ Or if you already have an API key, you can select **Manually enter API Key** and
 
 ---
 
+### Deep Infra
+
+1. Head over to the [Deep Infra dashboard](https://deepinfra.com/dash), create an account, and generate an API key.
+
+2. Run `opencode auth login` and select **Deep Infra**.
+
+   ```bash
+   $ opencode auth login
+
+   ┌  Add credential
+   │
+   ◆  Select provider
+   │  ● Deep Infra
+   │  ...
+   └
+   ```
+
+3. Enter your Deep Infra API key.
+
+   ```bash
+   $ opencode auth login
+
+   ┌  Add credential
+   │
+   ◇  Select provider
+   │  Deep Infra
+   │
+   ◇  Enter your API key
+   │  _
+   └
+   ```
+
+4. Run the `/models` command to select a model.
+
+---
+
 ### Fireworks AI
 
 1. Head over to the [Fireworks AI console](https://app.fireworks.ai/), create an account, and click **Create API Key**.
@@ -482,7 +518,6 @@ To use Google Vertex AI with OpenCode:
 
 4. Run the `/models` command to select a model like _Kimi-K2-Instruct_ or _GLM-4.6_.
 
-
 ---
 
 ### LM Studio
@@ -935,9 +970,9 @@ You can use any OpenAI-compatible provider with opencode. Most modern AI provide
 
 ##### Example
 
-Here's an example setting the `apiKey` and `headers` options.
+Here's an example setting the `apiKey`, `headers`, and model `limit` options.
 
-```json title="opencode.json" {9,11}
+```json title="opencode.json" {9,11,17-20}
 {
   "$schema": "https://opencode.ai/config.json",
   "provider": {
@@ -953,7 +988,11 @@ Here's an example setting the `apiKey` and `headers` options.
       },
       "models": {
         "my-model-name": {
-          "name": "My Model Display Name"
+          "name": "My Model Display Name",
+          "limit": {
+            "context": 200000,
+            "output": 65536
+          }
         }
       }
     }
@@ -961,7 +1000,14 @@ Here's an example setting the `apiKey` and `headers` options.
 }
 ```
 
-We are setting the `apiKey` using the `env` variable syntax, [learn more](/docs/config#env-vars).
+Configuration details:
+
+- **apiKey**: Set using `env` variable syntax, [learn more](/docs/config#env-vars).
+- **headers**: Custom headers sent with each request.
+- **limit.context**: Maximum input tokens the model accepts.
+- **limit.output**: Maximum tokens the model can generate.
+
+The `limit` fields allow OpenCode to understand how much context you have left. Standard providers pull these from models.dev automatically.
 
 ---
 
diff --git a/packages/web/src/content/docs/troubleshooting.mdx b/packages/web/src/content/docs/troubleshooting.mdx
index fb8e964ba..d38b892fc 100644
--- a/packages/web/src/content/docs/troubleshooting.mdx
+++ b/packages/web/src/content/docs/troubleshooting.mdx
@@ -90,8 +90,8 @@ If you encounter `ProviderModelNotFoundError` you are most likely incorrectly
 referencing a model somewhere.
 Models should be referenced like so: `<providerId>/<modelId>`
 
-
 Examples:
+
 - `openai/gpt-4.1`
 - `openrouter/google/gemini-2.5-flash`
 - `opencode/kimi-k2`
diff --git a/packages/web/src/content/docs/zen.mdx b/packages/web/src/content/docs/zen.mdx
index d2da4a51c..6c9b9fb88 100644
--- a/packages/web/src/content/docs/zen.mdx
+++ b/packages/web/src/content/docs/zen.mdx
@@ -75,6 +75,7 @@ You can also access our models through the following API endpoints.
 | Kimi K2           | kimi-k2           | `https://opencode.ai/zen/v1/chat/completions` | `@ai-sdk/openai-compatible` |
 | Qwen3 Coder 480B  | qwen3-coder       | `https://opencode.ai/zen/v1/chat/completions` | `@ai-sdk/openai-compatible` |
 | Grok Code Fast 1  | grok-code         | `https://opencode.ai/zen/v1/chat/completions` | `@ai-sdk/openai-compatible` |
+| Big Pickle        | big-pickle        | `https://opencode.ai/zen/v1/chat/completions` | `@ai-sdk/openai-compatible` |
 
 The [model id](/docs/config/#models) in your OpenCode config
 uses the format `opencode/<model-id>`. For example, for GPT 5 Codex, you would
@@ -117,8 +118,8 @@ We support a pay-as-you-go model. Below are the prices **per 1M tokens**.
 | GLM 4.6                           | $0.60  | $2.20  | $0.10       | -            |
 | Kimi K2                           | $0.60  | $2.50  | $0.36       | -            |
 | Qwen3 Coder 480B                  | $0.45  | $1.50  | -           | -            |
-| Grok Code Fast 1                  | Free   | Free   | -           | -            |
-| Code Supernova                    | Free   | Free   | -           | -            |
+| Grok Code Fast 1                  | Free   | Free   | Free        | -            |
+| Big Pickle                        | Free   | Free   | Free        | -            |
 | Claude Sonnet 4.5 (≤ 200K tokens) | $3.00  | $15.00 | $0.30       | $3.75        |
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50 | $0.60       | $7.50        |
 | Claude Sonnet 4 (≤ 200K tokens)   | $3.00  | $15.00 | $0.30       | $3.75        |
@@ -138,7 +139,7 @@ Credit card fees are passed along at cost; we don't charge anything beyond that.
 The free models:
 
 - Grok Code Fast 1 is currently free on OpenCode for a limited time. The xAI team is using this time to collect feedback and improve Grok Code.
-- Code Supernova is a stealth model that's free on OpenCode for a limited time. The team is using this time to collect feedback and improve the model.
+- Big Pickle is a stealth model that's free on OpenCode for a limited time. The team is using this time to collect feedback and improve the model.
 
 :::tip
 Subscription plans and a free tier are coming soon.
@@ -153,8 +154,7 @@ Subscription plans and a free tier are coming soon.
 All our models are hosted in the US. Our providers follow a zero-retention policy and do not use your data for model training, with the following exceptions:
 
 - Grok Code Fast 1: During its free period, collected data may be used to improve Grok Code.
-- Code Supernova: During its free period, collected data may be used to improve
-  the model.
+- Big Pickle: During its free period, collected data may be used to improve the model.
 - OpenAI APIs: Requests are retained for 30 days in accordance with [OpenAI's Data Policies](https://platform.openai.com/docs/guides/your-data).
 - Anthropic APIs: Requests are retained for 30 days in accordance with [Anthropic's Data Policies](https://docs.anthropic.com/en/docs/claude-code/data-usage).