diff --git a/packages/tui/src/feature-plugins/sidebar/context.tsx b/packages/tui/src/feature-plugins/sidebar/context.tsx
index f1c99d9679ce..02788f4f8c85 100644
--- a/packages/tui/src/feature-plugins/sidebar/context.tsx
+++ b/packages/tui/src/feature-plugins/sidebar/context.tsx
@@ -1,7 +1,8 @@
 import type { AssistantMessage } from "@opencode-ai/sdk/v2"
 import type { TuiPlugin, TuiPluginApi } from "@opencode-ai/plugin/tui"
 import type { BuiltinTuiPlugin } from "../builtins"
-import { createMemo } from "solid-js"
+import { createMemo, Show } from "solid-js"
+import { Locale } from "../../util/locale"
 
 const id = "internal:sidebar-context"
 
@@ -10,6 +11,22 @@ const money = new Intl.NumberFormat("en-US", {
   currency: "USD",
 })
 
+export function computeContextState(
+  last: AssistantMessage,
+  contextLimit?: number,
+) {
+  const tokens =
+    last.tokens.input + last.tokens.output + last.tokens.reasoning + last.tokens.cache.read + last.tokens.cache.write
+  const totalInput = last.tokens.input + last.tokens.cache.read
+  return {
+    tokens,
+    input: last.tokens.input,
+    cache: last.tokens.cache.read,
+    cachePercent: totalInput > 0 ? Math.round(last.tokens.cache.read / totalInput * 100) : 0,
+    percent: contextLimit ? Math.round((tokens / contextLimit) * 100) : null,
+  }
+}
+
 function View(props: { api: TuiPluginApi; session_id: string }) {
   const theme = () => props.api.theme.current
   const msg = createMemo(() => props.api.state.session.messages(props.session_id))
@@ -21,17 +38,15 @@ function View(props: { api: TuiPluginApi; session_id: string }) {
     if (!last) {
       return {
         tokens: 0,
+        input: 0,
+        cache: 0,
+        cachePercent: null,
         percent: null,
       }
     }
 
-    const tokens =
-      last.tokens.input + last.tokens.output + last.tokens.reasoning + last.tokens.cache.read + last.tokens.cache.write
     const model = props.api.state.provider.find((item) => item.id === last.providerID)?.models[last.modelID]
-    return {
-      tokens,
-      percent: model?.limit.context ? Math.round((tokens / model.limit.context) * 100) : null,
-    }
+    return computeContextState(last, model?.limit.context)
   })
 
   return (
@@ -39,8 +54,11 @@ function View(props: { api: TuiPluginApi; session_id: string }) {
       <text fg={theme().text}>
         <b>Context</b>
       </text>
-      <text fg={theme().textMuted}>{state().tokens.toLocaleString()} tokens</text>
-      <text fg={theme().textMuted}>{state().percent ?? 0}% used</text>
+      <text fg={theme().textMuted}>{state().tokens.toLocaleString()} tokens ({state().percent ?? 0}%)</text>
+      <Show when={state().input > 0 || state().cache > 0}>
+        <text fg={theme().textMuted}>Input {Locale.number(state().input)} · Cache {Locale.number(state().cache)}</text>
+        <text fg={theme().textMuted}>{state().cachePercent}% cached</text>
+      </Show>
       <text fg={theme().textMuted}>{money.format(cost())} spent</text>
     </box>
   )
diff --git a/packages/tui/test/feature-plugins/sidebar-context.test.ts b/packages/tui/test/feature-plugins/sidebar-context.test.ts
new file mode 100644
index 000000000000..7f8326b10070
--- /dev/null
+++ b/packages/tui/test/feature-plugins/sidebar-context.test.ts
@@ -0,0 +1,97 @@
+import { describe, expect, test } from "bun:test"
+import type { AssistantMessage } from "@opencode-ai/sdk/v2"
+import { computeContextState } from "../../src/feature-plugins/sidebar/context"
+
+function makeMessage(overrides: Partial<AssistantMessage["tokens"]> = {}): AssistantMessage {
+  return {
+    id: "msg_1",
+    sessionID: "ses_1",
+    role: "assistant",
+    agent: "build",
+    modelID: "claude-sonnet-4-20250514",
+    providerID: "anthropic",
+    mode: "",
+    parentID: "msg_0",
+    path: { cwd: "/test", root: "/test" },
+    cost: 0,
+    tokens: {
+      input: 0,
+      output: 0,
+      reasoning: 0,
+      cache: { read: 0, write: 0 },
+      ...overrides,
+    },
+    time: { created: 0 },
+  }
+}
+
+describe("sidebar context cache calculation", () => {
+  test("returns 0% cache when no cache read tokens", () => {
+    const msg = makeMessage({ input: 1000, output: 500, cache: { read: 0, write: 0 } })
+    const state = computeContextState(msg)
+    expect(state.cachePercent).toBe(0)
+    expect(state.input).toBe(1000)
+    expect(state.cache).toBe(0)
+  })
+
+  test("calculates cache hit percentage correctly", () => {
+    const msg = makeMessage({ input: 800, output: 500, cache: { read: 200, write: 0 } })
+    const state = computeContextState(msg)
+    expect(state.input).toBe(800)
+    expect(state.cache).toBe(200)
+    expect(state.cachePercent).toBe(20)
+  })
+
+  test("100% cache hit when all input is cached", () => {
+    const msg = makeMessage({ input: 0, output: 500, cache: { read: 1000, write: 0 } })
+    const state = computeContextState(msg)
+    expect(state.input).toBe(0)
+    expect(state.cache).toBe(1000)
+    expect(state.cachePercent).toBe(100)
+  })
+
+  test("does not include cache.write in cache hit calculation", () => {
+    const msg = makeMessage({ input: 800, output: 500, cache: { read: 200, write: 500 } })
+    const state = computeContextState(msg)
+    expect(state.cache).toBe(200)
+    expect(state.cachePercent).toBe(20)
+  })
+
+  test("cache.write is included in total token count", () => {
+    const msg = makeMessage({ input: 800, output: 500, reasoning: 100, cache: { read: 200, write: 400 } })
+    const state = computeContextState(msg)
+    expect(state.tokens).toBe(800 + 500 + 100 + 200 + 400)
+  })
+
+  test("handles zero input and zero cache gracefully", () => {
+    const msg = makeMessage({ input: 0, output: 0, cache: { read: 0, write: 0 } })
+    const state = computeContextState(msg)
+    expect(state.cachePercent).toBe(0)
+    expect(state.input).toBe(0)
+    expect(state.cache).toBe(0)
+  })
+
+  test("rounds cache percentage to nearest integer", () => {
+    const msg = makeMessage({ input: 1000, output: 500, cache: { read: 333, write: 0 } })
+    const state = computeContextState(msg)
+    expect(state.cachePercent).toBe(25)
+  })
+
+  test("50% cache hit", () => {
+    const msg = makeMessage({ input: 500, output: 500, cache: { read: 500, write: 0 } })
+    const state = computeContextState(msg)
+    expect(state.cachePercent).toBe(50)
+  })
+
+  test("calculates context percentage when limit provided", () => {
+    const msg = makeMessage({ input: 1000, output: 500, cache: { read: 0, write: 0 } })
+    const state = computeContextState(msg, 200000)
+    expect(state.percent).toBe(1)
+  })
+
+  test("returns null percent when no context limit", () => {
+    const msg = makeMessage({ input: 1000, output: 500 })
+    const state = computeContextState(msg)
+    expect(state.percent).toBeNull()
+  })
+})