diff --git a/packages/core/test/config/provider.test.ts b/packages/core/test/config/provider.test.ts
index d69881656ee..b06dccc75b7 100644
--- a/packages/core/test/config/provider.test.ts
+++ b/packages/core/test/config/provider.test.ts
@@ -77,7 +77,7 @@ describe("ConfigProviderPlugin.Plugin", () => {
     }),
   )
 
-  it.effect("defaults custom models to agent capabilities", () =>
+  it.effect("defaults custom model metadata", () =>
     Effect.gen(function* () {
       const catalog = yield* Catalog.Service
       const providerID = Provider.ID.make("custom")
@@ -100,6 +100,7 @@ describe("ConfigProviderPlugin.Plugin", () => {
 
       const model = required(yield* catalog.model.get(providerID, modelID))
       expect(model.capabilities).toEqual({ tools: true, input: ["text", "image"], output: ["text"] })
+      expect(model.limit).toEqual({ context: 200_000, output: 32_000 })
     }),
   )
 
diff --git a/packages/schema/src/model.ts b/packages/schema/src/model.ts
index a1e396f19e8..cef4b39c6df 100644
--- a/packages/schema/src/model.ts
+++ b/packages/schema/src/model.ts
@@ -125,7 +125,7 @@ export const Info = Schema.Struct({
           cost: [],
           status: "active",
           enabled: true,
-          limit: { context: 0, output: 0 },
+          limit: { context: 200_000, output: 32_000 },
         }) satisfies Info,
     })),
   )
diff --git a/packages/schema/test/model.test.ts b/packages/schema/test/model.test.ts
index 889f2e63896..642e9322985 100644
--- a/packages/schema/test/model.test.ts
+++ b/packages/schema/test/model.test.ts
@@ -1,6 +1,7 @@
 import { describe, expect, test } from "bun:test"
 import { Schema } from "effect"
 import { Model } from "../src/model.js"
+import { Provider } from "../src/provider.js"
 
 describe("Model.Ref", () => {
   test("parses model references with optional variants", () => {
@@ -49,3 +50,11 @@ describe("Model.Compatibility", () => {
     })
   })
 })
+
+describe("Model.Info", () => {
+  test("uses practical token limits for unknown models", () => {
+    const model = Model.Info.default(Provider.ID.make("custom"), Model.ID.make("gpt-5.6"))
+
+    expect(model.limit).toEqual({ context: 200_000, output: 32_000 })
+  })
+})
diff --git a/packages/www/content/docs/(Configure)/models.mdx b/packages/www/content/docs/(Configure)/models.mdx
index 1ecb2f90a39..6b6d757a391 100644
--- a/packages/www/content/docs/(Configure)/models.mdx
+++ b/packages/www/content/docs/(Configure)/models.mdx
@@ -85,9 +85,14 @@ You can also map a friendly catalog ID to a different API model ID with `modelID
 ```
 
 Here `openai/coding-default` is the selectable catalog reference, while `gpt-5.2` is sent to the provider. A model that is
-not already in the catalog defaults to tool support, text and image input, and text output. Set accurate `capabilities`
-and `limit` values when those defaults do not match the model or OpenCode needs to enforce its context limits. Set
-`disabled: true` on a model entry to hide it from the available catalog.
+not already in the catalog receives fallback metadata:
+
+- Tool support, text and image input, and text output.
+- A 200,000-token context limit and 32,000-token output limit. The input limit remains unspecified.
+
+These values are assumptions, not model discovery. Configure accurate `capabilities` and `limit` values whenever they are
+known; explicit values override the fallbacks. Set `disabled: true` on a model entry to hide it from the available
+catalog.
 
 OpenAI-compatible models that stream reasoning through a custom assistant-message field can set
 `compatibility.reasoningField`: