feat(zai): add GLM-5.2 support

.env.example

@@ -183,8 +183,12 @@ ANTHROPIC_API_KEY=sk-ant-your-key-here
 # Legacy aliases also work: deepseek-chat and deepseek-reasoner
 # For Z.AI GLM Coding Plan, set:
 # OPENAI_BASE_URL=https://api.z.ai/api/coding/paas/v4
-# OPENAI_MODEL=GLM-5.1
-# Optional: OPENAI_MODEL=GLM-5-Turbo, GLM-4.7, or GLM-4.5-Air
+# OPENAI_MODEL=glm-5.2
+# Optional: OPENAI_MODEL=GLM-5.1, GLM-5-Turbo, GLM-4.7, or GLM-4.5-Air
+# Optional GLM-5.2 thinking controls:
+# OPENAI_MODEL='glm-5.2?reasoning=high'     # enhanced reasoning
+# OPENAI_MODEL='glm-5.2?reasoning=xhigh'    # maps to Z.AI reasoning_effort=max
+# OPENAI_MODEL='glm-5.2?thinking=disabled'  # faster direct answers for simple tasks
 # For Hicap, use the OpenAI-compatible route flag above and set:
 # HICAP_API_KEY=your-hicap-key-here
 # OPENAI_BASE_URL=https://api.hicap.ai/v1

README.md

@@ -105,7 +105,7 @@ Inside OpenClaude:
 - run `/provider` for guided provider setup and saved profiles
 - run `/onboard-github` for GitHub Models onboarding
 
-> **Note:** OpenClaude does not automatically load project `.env` files. We recommend using the `/provider` command for setup, which securely stores credentials. If you prefer environment variables, export them explicitly or run `openclaude --provider-env-file .env` for provider/setup variables. Export runtime/debug knobs from your shell or launcher.
+> **Note:** OpenClaude does not automatically load project `.env` files. We recommend using the `/provider` command for setup, which saves provider profiles and credentials in `.openclaude-profile.json`. If you prefer environment variables, export them explicitly or run `openclaude --provider-env-file .env` for provider/setup variables. Export runtime/debug knobs from your shell or launcher.
 
 ### Background sessions
 
@@ -195,6 +195,7 @@ Advanced and source-build guides:
 | Provider | Setup Path | Notes |
 | --- | --- | --- |
 | OpenAI-compatible | `/provider` or env vars | Works with OpenAI, OpenRouter, DeepSeek, Groq, Mistral, LM Studio, and other compatible `/v1` servers |
+| Z.AI GLM Coding Plan | `/provider` or OpenAI-compatible env vars | Uses `OPENAI_API_KEY` at `https://api.z.ai/api/coding/paas/v4` and defaults to `glm-5.2` |
 | Hicap | `/provider` or OpenAI-compatible env vars | Uses `api-key` auth, discovers models from unauthenticated `/models`, and supports Responses mode for `gpt-` models |
 | Fireworks AI | `/provider` or env vars | First-class provider with 276 curated models (DeepSeek, Qwen, Llama, Gemma, and more); uses `FIREWORKS_API_KEY` |
 | Gemini | `/provider` or env vars | Supports API key only |
@@ -228,6 +229,7 @@ OpenClaude supports multiple providers, but behavior is not identical across all
 - Smaller local models can struggle with long multi-step tool flows
 - Some providers impose lower output caps than the CLI defaults, and OpenClaude adapts where possible
 - Gitlawb Opengateway is the fresh-install startup default and requires an API key from https://gitlawb.com/opengateway/keys. It uses one OpenAI-compatible base URL; switch between `mimo-*` and `google/gemini-3.1-flash-lite-preview` with `/model`, and do not pin the base URL to `/v1/xiaomi-mimo`.
+- Z.AI GLM Coding Plan uses `https://api.z.ai/api/coding/paas/v4` with `glm-5.2` by default. Use `glm-5.2?reasoning=high` for enhanced reasoning, `glm-5.2?reasoning=xhigh` to request Z.AI `reasoning_effort=max`, or `glm-5.2?thinking=disabled` for faster direct answers.
 - Xiaomi MiMo uses `api-key` header auth on the direct OpenAI-compatible route and currently does not support `/usage` reporting in OpenClaude
 
 ### GitHub Copilot sub-agent optimization
@@ -261,7 +263,7 @@ Add to `~/.openclaude.json`:
       "api_key": "sk-your-key"
     },
     "zai-default": {
-      "model": "glm-5.1",
+      "model": "glm-5.2",
       "base_url": "https://api.z.ai/api/coding/paas/v4",
       "api_key": "sk-your-key"
     },
@@ -282,7 +284,7 @@ Add to `~/.openclaude.json`:
 
 When no routing match is found, the global provider remains the fallback.
 
-`agentRouting` values and explicit Agent tool `model` overrides match keys in `agentModels`. By default, that key is also the model string sent to the provider. Set `agentModels.<key>.model` when you want a local route key such as `zai-default` to call a different provider model name such as `glm-5.1`.
+`agentRouting` values and explicit Agent tool `model` overrides match keys in `agentModels`. By default, that key is also the model string sent to the provider. Set `agentModels.<key>.model` when you want a local route key such as `zai-default` to call a different provider model name such as `glm-5.2`.
 
 > **Note:** `/provider` changes the global/parent provider for your current session. `agentModels` and `agentRouting` are specifically for configuring per-agent provider overrides while keeping the parent session unchanged.
 

src/integrations/brands/glm.ts

@@ -13,6 +13,7 @@ export default defineBrand({
     supportsPreciseTokenCount: false,
   },
   modelIds: [
+    'glm-5.2',
     'GLM-5.1',
     'GLM-5-Turbo',
     'GLM-5',

src/integrations/descriptors.ts

@@ -37,7 +37,8 @@ export interface OpenAIShimTransportConfig {
   preserveReasoningContent?: boolean
   requireReasoningContentOnAssistantMessages?: boolean
   reasoningContentFallback?: '' | 'omit'
-  thinkingRequestFormat?: 'none' | 'deepseek-compatible'
+  thinkingRequestFormat?: 'none' | 'deepseek-compatible' | 'zai-compatible'
+  enableToolStreaming?: boolean
   maxTokensField?: OpenAIShimTokenField
   removeBodyFields?: string[]
   /** Override the endpoint path for this model (e.g., '/responses', '/messages'). */

src/integrations/models/glm.ts

@@ -29,6 +29,7 @@ function glmModel(
 }
 
 export default [
+  glmModel('glm-5.2', 'GLM 5.2', 1_000_000, 131_072),
   glmModel('GLM-5.1', 'GLM-5.1', 202_752, 131_072),
   glmModel('GLM-5-Turbo', 'GLM-5-Turbo', 202_752, 131_072),
   glmModel('GLM-5', 'GLM-5', 202_752, 131_072),

src/integrations/runtimeMetadata.test.ts

@@ -70,6 +70,39 @@ describe('resolveModelRuntimeLimits', () => {
       ).toBe(1_000_000)
     })
   })
+
+  it('uses built-in Z.AI GLM-5.2 runtime limits', () => {
+    const limits = resolveModelRuntimeLimits({
+      model: 'glm-5.2',
+      processEnv: {
+        OPENAI_BASE_URL: 'https://api.z.ai/api/coding/paas/v4',
+      },
+    })
+
+    expect(limits.contextWindow).toBe(1_000_000)
+    expect(limits.maxOutputTokens).toBe(131_072)
+  })
+})
+
+describe('resolveOpenAIShimRuntimeContext - Z.AI GLM-5.2', () => {
+  it.each([
+    'glm-5.2',
+    'glm-5.2?reasoning=high',
+    'glm-5.2?thinking=disabled',
+  ])('uses Z.AI GLM-5.2 shim settings for %s', model => {
+    const result = resolveOpenAIShimRuntimeContext({
+      model,
+      baseUrl: 'https://api.z.ai/api/coding/paas/v4',
+      processEnv: {},
+    })
+
+    expect(result.routeId).toBe('zai')
+    expect(result.catalogEntry?.id).toBe('glm-5.2')
+    expect(result.openaiShimConfig.thinkingRequestFormat).toBe('zai-compatible')
+    expect(result.openaiShimConfig.preserveReasoningContent).toBe(true)
+    expect(result.openaiShimConfig.requireReasoningContentOnAssistantMessages).toBe(true)
+    expect(result.openaiShimConfig.enableToolStreaming).toBe(true)
+  })
 })
 
 describe('resolveOpenAIShimRuntimeContext - segment-boundary heuristic', () => {

src/integrations/runtimeMetadata.ts

@@ -29,8 +29,20 @@ import { parseCustomHeadersEnv } from '../utils/providerCustomHeaders.js'
 function normalizeModelApiName(
   value: string | undefined,
 ): string | null {
-  const trimmed = value?.trim().toLowerCase()
-  return trimmed ? trimmed : null
+  const baseModel = getBaseModelApiName(value)
+  return baseModel ? baseModel.toLowerCase() : null
+}
+
+function getBaseModelApiName(value: string | undefined): string | null {
+  const trimmed = value?.trim()
+  if (!trimmed) {
+    return null
+  }
+
+  const queryIndex = trimmed.indexOf('?')
+  const baseModel =
+    queryIndex === -1 ? trimmed : trimmed.slice(0, queryIndex).trim()
+  return baseModel || null
 }
 
 function matchesCatalogEntryModel(
@@ -269,7 +281,7 @@ function findModelDescriptorForApiName(
   routeId: string | null,
   modelApiName: string | undefined,
 ) {
-  const trimmedModel = modelApiName?.trim()
+  const trimmedModel = getBaseModelApiName(modelApiName)
   if (!trimmedModel) {
     return null
   }
@@ -385,22 +397,23 @@ export function resolveModelRuntimeLimits(options: {
   const routeId = resolveActiveRouteIdFromEnv(runtimeEnv, {
     activeProfileProvider: options.activeProfileProvider,
   })
-  const catalogEntry = findCatalogEntryForApiName(routeId, options.model)
+  const modelApiName = getBaseModelApiName(options.model) ?? options.model
+  const catalogEntry = findCatalogEntryForApiName(routeId, modelApiName)
   const cachedCatalogEntry = findCachedCatalogEntryForApiName(
     routeId,
-    options.model,
+    modelApiName,
     runtimeEnv,
   )
   const modelDescriptor =
     getModelDescriptorForCatalogEntry(catalogEntry) ??
     getModelDescriptorForCatalogEntry(cachedCatalogEntry) ??
-    findModelDescriptorForApiName(routeId, options.model)
+    findModelDescriptorForApiName(routeId, modelApiName)
   const externalContextWindow = getOpenAIContextWindowMatches(
-    options.model,
+    modelApiName,
     runtimeEnv,
   )
   const externalMaxOutputTokens = getOpenAIMaxOutputTokenMatches(
-    options.model,
+    modelApiName,
     runtimeEnv,
   )
 

src/integrations/vendors/zai.ts

@@ -5,7 +5,7 @@ export default defineVendor({
   label: 'Z.AI',
   classification: 'openai-compatible',
   defaultBaseUrl: 'https://api.z.ai/api/coding/paas/v4',
-  defaultModel: 'GLM-5.1',
+  defaultModel: 'glm-5.2',
   requiredEnvVars: ['OPENAI_API_KEY'],
   setup: {
     requiresAuth: true,
@@ -18,7 +18,7 @@ export default defineVendor({
       preserveReasoningContent: true,
       requireReasoningContentOnAssistantMessages: true,
       reasoningContentFallback: '',
-      thinkingRequestFormat: 'deepseek-compatible',
+      thinkingRequestFormat: 'zai-compatible',
       maxTokensField: 'max_tokens',
       removeBodyFields: ['store'],
     },
@@ -44,6 +44,17 @@ export default defineVendor({
   catalog: {
     source: 'static',
     models: [
+      {
+        id: 'glm-5.2',
+        apiName: 'glm-5.2',
+        label: 'GLM-5.2',
+        modelDescriptorId: 'glm-5.2',
+        transportOverrides: {
+          openaiShim: {
+            enableToolStreaming: true,
+          },
+        },
+      },
       {
         id: 'GLM-5.1',
         apiName: 'GLM-5.1',