tweak: move the max token exclusions to plugins @rekram1-node (#21225)

Author: rekram1-node

packages/opencode/src/plugin/codex.ts

@@ -599,5 +599,10 @@ export async function CodexAuthPlugin(input: PluginInput): Promise<Hooks> {
       output.headers["User-Agent"] = `opencode/${Installation.VERSION} (${os.platform()} ${os.release()}; ${os.arch()})`
       output.headers.session_id = input.sessionID
     },
+    "chat.params": async (input, output) => {
+      if (input.model.providerID !== "openai") return
+      // Match codex cli
+      output.maxOutputTokens = undefined
+    },
   }
 }

packages/opencode/src/plugin/github-copilot/copilot.ts

@@ -309,6 +309,14 @@ export async function CopilotAuthPlugin(input: PluginInput): Promise<Hooks> {
         },
       ],
     },
+    "chat.params": async (incoming, output) => {
+      if (!incoming.model.providerID.includes("github-copilot")) return
+
+      // Match github copilot cli, omit maxOutputTokens for gpt models
+      if (incoming.model.api.id.includes("gpt")) {
+        output.maxOutputTokens = undefined
+      }
+    },
     "chat.headers": async (incoming, output) => {
       if (!incoming.model.providerID.includes("github-copilot")) return
 

packages/opencode/src/session/llm.ts

@@ -160,11 +160,6 @@ export namespace LLM {
             ...input.messages,
           ]
 
-    const maxOutputTokens =
-      isOpenaiOauth || provider.id.includes("github-copilot")
-        ? undefined
-        : ProviderTransform.maxOutputTokens(input.model)
-
     const params = await Plugin.trigger(
       "chat.params",
       {
@@ -180,7 +175,7 @@ export namespace LLM {
           : undefined,
         topP: input.agent.topP ?? ProviderTransform.topP(input.model),
         topK: ProviderTransform.topK(input.model),
-        maxOutputTokens,
+        maxOutputTokens: ProviderTransform.maxOutputTokens(input.model),
         options,
       },
     )

packages/opencode/test/session/llm.test.ts

@@ -743,8 +743,7 @@ describe("session.llm.stream", () => {
         expect((body.reasoning as { effort?: string } | undefined)?.effort).toBe("high")
 
         const maxTokens = body.max_output_tokens as number | undefined
-        const expectedMaxTokens = ProviderTransform.maxOutputTokens(resolved)
-        expect(maxTokens).toBe(expectedMaxTokens)
+        expect(maxTokens).toBe(undefined) // match codex cli behavior
       },
     })
   })