feat(cache): add optional 1h TTL on first system cache marker behind OPENCODE_EXPERIMENTAL_CACHE_1H_TTL flag

Author: bhagirathsinh-vaghela

packages/opencode/src/flag/flag.ts

@@ -59,6 +59,7 @@ export namespace Flag {
   export const OPENCODE_EXPERIMENTAL_PLAN_MODE = OPENCODE_EXPERIMENTAL || truthy("OPENCODE_EXPERIMENTAL_PLAN_MODE")
   export const OPENCODE_EXPERIMENTAL_MARKDOWN = !falsy("OPENCODE_EXPERIMENTAL_MARKDOWN")
   export const OPENCODE_EXPERIMENTAL_CACHE_STABILIZATION = truthy("OPENCODE_EXPERIMENTAL_CACHE_STABILIZATION")
+  export const OPENCODE_EXPERIMENTAL_CACHE_1H_TTL = truthy("OPENCODE_EXPERIMENTAL_CACHE_1H_TTL")
   export const OPENCODE_MODELS_URL = process.env["OPENCODE_MODELS_URL"]
   export const OPENCODE_MODELS_PATH = process.env["OPENCODE_MODELS_PATH"]
 

packages/opencode/src/provider/transform.ts

@@ -171,10 +171,12 @@ export namespace ProviderTransform {
     return msgs
   }
 
-  function applyCaching(msgs: ModelMessage[], model: Provider.Model): ModelMessage[] {
+  function applyCaching(msgs: ModelMessage[], model: Provider.Model, extendedTTL?: boolean): ModelMessage[] {
     const system = msgs.filter((msg) => msg.role === "system").slice(0, 2)
     const final = msgs.filter((msg) => msg.role !== "system").slice(-2)
 
+    // Use 1h cache TTL on first system block (2x write cost vs 1.25x for default 5-min)
+    const anthropicCache = extendedTTL ? { type: "ephemeral", ttl: "1h" } : { type: "ephemeral" }
     const providerOptions = {
       anthropic: {
         cacheControl: { type: "ephemeral" },
@@ -194,18 +196,21 @@ export namespace ProviderTransform {
     }
 
     for (const msg of unique([...system, ...final])) {
+      const options = msg === system[0]
+        ? { ...providerOptions, anthropic: { cacheControl: anthropicCache } }
+        : providerOptions
       const useMessageLevelOptions = model.providerID === "anthropic" || model.providerID.includes("bedrock")
       const shouldUseContentOptions = !useMessageLevelOptions && Array.isArray(msg.content) && msg.content.length > 0
 
       if (shouldUseContentOptions) {
         const lastContent = msg.content[msg.content.length - 1]
         if (lastContent && typeof lastContent === "object") {
-          lastContent.providerOptions = mergeDeep(lastContent.providerOptions ?? {}, providerOptions)
+          lastContent.providerOptions = mergeDeep(lastContent.providerOptions ?? {}, options)
           continue
         }
       }
 
-      msg.providerOptions = mergeDeep(msg.providerOptions ?? {}, providerOptions)
+      msg.providerOptions = mergeDeep(msg.providerOptions ?? {}, options)
     }
 
     return msgs
@@ -261,7 +266,7 @@ export namespace ProviderTransform {
         model.api.npm === "@ai-sdk/anthropic") &&
       model.api.npm !== "@ai-sdk/gateway"
     ) {
-      msgs = applyCaching(msgs, model)
+      msgs = applyCaching(msgs, model, (options.extendedTTL as boolean) ?? Flag.OPENCODE_EXPERIMENTAL_CACHE_1H_TTL)
     }
 
     // Remap providerOptions keys from stored providerID to expected SDK key

packages/opencode/test/provider/transform.test.ts

@@ -1531,6 +1531,82 @@ describe("ProviderTransform.message - claude w/bedrock custom inference profile"
 })
 
 
+describe("ProviderTransform.message - first system block gets 1h TTL when flag set", () => {
+  const anthropicModel = {
+    id: "anthropic/claude-sonnet-4-6",
+    providerID: "anthropic",
+    api: {
+      id: "claude-sonnet-4-6",
+      url: "https://api.anthropic.com",
+      npm: "@ai-sdk/anthropic",
+    },
+    capabilities: {
+      temperature: true,
+      reasoning: false,
+      attachment: true,
+      toolcall: true,
+      input: { text: true, audio: false, image: true, video: false, pdf: true },
+      output: { text: true, audio: false, image: false, video: false, pdf: false },
+      interleaved: false,
+    },
+    cost: { input: 0.003, output: 0.015, cache: { read: 0.0003, write: 0.00375 } },
+    limit: { context: 200000, output: 8192 },
+    status: "active",
+    options: {},
+    headers: {},
+  } as any
+
+  test("first system block gets 1h TTL when extendedTTL is true", () => {
+    const msgs = [
+      { role: "system", content: "Block 1" },
+      { role: "system", content: "Block 2" },
+      { role: "user", content: "Hello" },
+    ] as any[]
+
+    const result = ProviderTransform.message(msgs, anthropicModel, { extendedTTL: true }) as any[]
+
+    expect(result[0].providerOptions.anthropic.cacheControl).toEqual({ type: "ephemeral", ttl: "1h" })
+  })
+
+  test("first system block gets default ephemeral when extendedTTL is not set", () => {
+    const msgs = [
+      { role: "system", content: "Block 1" },
+      { role: "system", content: "Block 2" },
+      { role: "user", content: "Hello" },
+    ] as any[]
+
+    const result = ProviderTransform.message(msgs, anthropicModel, {}) as any[]
+
+    expect(result[0].providerOptions.anthropic.cacheControl).toEqual({ type: "ephemeral" })
+  })
+
+  test("second system block always gets default ephemeral TTL", () => {
+    const msgs = [
+      { role: "system", content: "Block 1" },
+      { role: "system", content: "Block 2" },
+      { role: "user", content: "Hello" },
+    ] as any[]
+
+    const result = ProviderTransform.message(msgs, anthropicModel, { extendedTTL: true }) as any[]
+
+    expect(result[1].providerOptions.anthropic.cacheControl).toEqual({ type: "ephemeral" })
+  })
+
+  test("conversation messages get default ephemeral TTL even with extendedTTL", () => {
+    const msgs = [
+      { role: "system", content: "System" },
+      { role: "user", content: "Hello" },
+      { role: "assistant", content: "Hi" },
+      { role: "user", content: "World" },
+    ] as any[]
+
+    const result = ProviderTransform.message(msgs, anthropicModel, { extendedTTL: true }) as any[]
+
+    const last = result[result.length - 1]
+    expect(last.providerOptions.anthropic.cacheControl).toEqual({ type: "ephemeral" })
+  })
+})
+
 describe("ProviderTransform.message - cache control on gateway", () => {
   const createModel = (overrides: Partial<any> = {}) =>
     ({