Skip to content

Add zai-coding-plan provider and make zai point to regular API #899

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
kaofelix wants to merge 10 commits into
base: main
Choose a base branch
from
Open

Add zai-coding-plan provider and make zai point to regular API #899

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
b483a11
feat(ai): add zai-coding-plan provider and make zai point to regular API
kaofelix Jan 22, 2026
fc0d906
docs(coding-agent): add Z.AI provider documentation
assagman Jan 20, 2026
3344add
replace zai with zai-coding-plan
kaofelix Jan 22, 2026
1b924f2
simplify documentation for zai coding plan
kaofelix Jan 22, 2026
7034ebe
update changelog
kaofelix Jan 22, 2026
f5b972e
solve readme conflict
kaofelix Jan 25, 2026
1732777
Merge remote-tracking branch 'upstream/main' into zai-coding-plan-pro…
kaofelix Jan 26, 2026
02893b7
move the zai provider change to breaking changes again
kaofelix Jan 26, 2026
ecb583d
Merge branch 'main' into zai-coding-plan-provider
kaofelix Jan 26, 2026
98ddd86
Merge branch 'main' into zai-coding-plan-provider
kaofelix Jan 26, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions packages/agent/test/e2e.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -302,8 +302,8 @@ describe("Agent E2E Tests", () => {
});
});

describe.skipIf(!process.env.ZAI_API_KEY)("zAI Provider (glm-4.5-air)", () => {
const model = getModel("zai", "glm-4.5-air");
describe.skipIf(!process.env.ZAI_API_KEY)("zAI Coding Plan Provider (glm-4.5-air)", () => {
const model = getModel("zai-coding-plan", "glm-4.5-air");

it("should handle basic text prompt", async () => {
await basicPrompt(model);
Expand Down
3 changes: 3 additions & 0 deletions packages/ai/CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,9 @@

## [Unreleased]

### Breaking Changes

- Renamed `zai` provider to `zai-coding-plan` to align with naming in models.dev and be explicit about which endpoint we use
## [0.50.1] - 2026-01-26

### Fixed
Expand Down
3 changes: 2 additions & 1 deletion packages/ai/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,6 +56,7 @@ Unified LLM API with automatic model discovery, provider configuration, token an
- **Groq**
- **Cerebras**
- **xAI**
- **zAI Coding Plan**
- **OpenRouter**
- **Vercel AI Gateway**
- **MiniMax**
Expand Down Expand Up @@ -892,7 +893,7 @@ In Node.js environments, you can set environment variables to avoid passing API
| xAI | `XAI_API_KEY` |
| OpenRouter | `OPENROUTER_API_KEY` |
| Vercel AI Gateway | `AI_GATEWAY_API_KEY` |
| zAI | `ZAI_API_KEY` |
| zAI Coding Plan | `ZAI_API_KEY` |
| MiniMax | `MINIMAX_API_KEY` |
| GitHub Copilot | `COPILOT_GITHUB_TOKEN` or `GH_TOKEN` or `GITHUB_TOKEN` |

Expand Down
44 changes: 22 additions & 22 deletions packages/ai/scripts/generate-models.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -417,33 +417,33 @@ async function loadModelsDevData(): Promise<Model<any>[]> {
}
}

// Process zAi models
if (data.zai?.models) {
for (const [modelId, model] of Object.entries(data.zai.models)) {
// Process zAi Coding Plan models
if (data["zai-coding-plan"]?.models) {
for (const [modelId, model] of Object.entries(data["zai-coding-plan"].models)) {
const m = model as ModelsDevModel;
if (m.tool_call !== true) continue;
const supportsImage = m.modalities?.input?.includes("image")

models.push({
id: modelId,
name: m.name || modelId,
api: "openai-completions",
provider: "zai",
baseUrl: "https://api.z.ai/api/coding/paas/v4",
reasoning: m.reasoning === true,
input: supportsImage ? ["text", "image"] : ["text"],
cost: {
input: m.cost?.input || 0,
output: m.cost?.output || 0,
cacheRead: m.cost?.cache_read || 0,
cacheWrite: m.cost?.cache_write || 0,
},
compat: {
supportsDeveloperRole: false,
thinkingFormat: "zai",
},
contextWindow: m.limit?.context || 4096,
maxTokens: m.limit?.output || 4096,
id: modelId,
name: m.name || modelId,
api: "openai-completions",
provider: "zai-coding-plan",
baseUrl: "https://api.z.ai/api/coding/paas/v4",
reasoning: m.reasoning === true,
input: supportsImage ? ["text", "image"] : ["text"],
cost: {
input: m.cost?.input || 0,
output: m.cost?.output || 0,
cacheRead: m.cost?.cache_read || 0,
cacheWrite: m.cost?.cache_write || 0,
},
compat: {
supportsDeveloperRole: false,
thinkingFormat: "zai",
},
contextWindow: m.limit?.context || 4096,
maxTokens: m.limit?.output || 4096,
});
}
}
Expand Down
1 change: 1 addition & 0 deletions packages/ai/src/env-api-keys.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -102,6 +102,7 @@ export function getEnvApiKey(provider: any): string | undefined {
openrouter: "OPENROUTER_API_KEY",
"vercel-ai-gateway": "AI_GATEWAY_API_KEY",
zai: "ZAI_API_KEY",
"zai-coding-plan": "ZAI_API_KEY",
mistral: "MISTRAL_API_KEY",
minimax: "MINIMAX_API_KEY",
"minimax-cn": "MINIMAX_CN_API_KEY",
Expand Down
68 changes: 43 additions & 25 deletions packages/ai/src/models.generated.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4886,7 +4886,7 @@ export const MODELS = {
input: ["text"],
cost: {
input: 0.09,
output: 0.39999999999999997,
output: 0.44999999999999996,
cacheRead: 0,
cacheWrite: 0,
},
Expand Down Expand Up @@ -11171,20 +11171,20 @@ export const MODELS = {
maxTokens: 4096,
} satisfies Model<"openai-completions">,
},
"zai": {
"zai-coding-plan": {
"glm-4.5": {
id: "glm-4.5",
name: "GLM-4.5",
api: "openai-completions",
provider: "zai",
provider: "zai-coding-plan",
baseUrl: "https://api.z.ai/api/coding/paas/v4",
compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
reasoning: true,
input: ["text"],
cost: {
input: 0.6,
output: 2.2,
cacheRead: 0.11,
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
contextWindow: 131072,
Expand All @@ -11194,15 +11194,15 @@ export const MODELS = {
id: "glm-4.5-air",
name: "GLM-4.5-Air",
api: "openai-completions",
provider: "zai",
provider: "zai-coding-plan",
baseUrl: "https://api.z.ai/api/coding/paas/v4",
compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
reasoning: true,
input: ["text"],
cost: {
input: 0.2,
output: 1.1,
cacheRead: 0.03,
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
contextWindow: 131072,
Expand All @@ -11212,7 +11212,7 @@ export const MODELS = {
id: "glm-4.5-flash",
name: "GLM-4.5-Flash",
api: "openai-completions",
provider: "zai",
provider: "zai-coding-plan",
baseUrl: "https://api.z.ai/api/coding/paas/v4",
compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
reasoning: true,
Expand All @@ -11230,14 +11230,14 @@ export const MODELS = {
id: "glm-4.5v",
name: "GLM-4.5V",
api: "openai-completions",
provider: "zai",
provider: "zai-coding-plan",
baseUrl: "https://api.z.ai/api/coding/paas/v4",
compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
reasoning: true,
input: ["text", "image"],
cost: {
input: 0.6,
output: 1.8,
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
Expand All @@ -11248,15 +11248,15 @@ export const MODELS = {
id: "glm-4.6",
name: "GLM-4.6",
api: "openai-completions",
provider: "zai",
provider: "zai-coding-plan",
baseUrl: "https://api.z.ai/api/coding/paas/v4",
compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
reasoning: true,
input: ["text"],
cost: {
input: 0.6,
output: 2.2,
cacheRead: 0.11,
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
contextWindow: 204800,
Expand All @@ -11266,14 +11266,14 @@ export const MODELS = {
id: "glm-4.6v",
name: "GLM-4.6V",
api: "openai-completions",
provider: "zai",
provider: "zai-coding-plan",
baseUrl: "https://api.z.ai/api/coding/paas/v4",
compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
reasoning: true,
input: ["text", "image"],
cost: {
input: 0.3,
output: 0.9,
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
Expand All @@ -11284,19 +11284,37 @@ export const MODELS = {
id: "glm-4.7",
name: "GLM-4.7",
api: "openai-completions",
provider: "zai",
provider: "zai-coding-plan",
baseUrl: "https://api.z.ai/api/coding/paas/v4",
compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
reasoning: true,
input: ["text"],
cost: {
input: 0.6,
output: 2.2,
cacheRead: 0.11,
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
contextWindow: 204800,
maxTokens: 131072,
} satisfies Model<"openai-completions">,
"glm-4.7-flash": {
id: "glm-4.7-flash",
name: "GLM-4.7-Flash",
api: "openai-completions",
provider: "zai-coding-plan",
baseUrl: "https://api.z.ai/api/coding/paas/v4",
compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
reasoning: true,
input: ["text"],
cost: {
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
contextWindow: 200000,
maxTokens: 131072,
} satisfies Model<"openai-completions">,
},
} as const;
36 changes: 29 additions & 7 deletions packages/ai/src/providers/openai-completions.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -105,7 +105,9 @@ export const streamOpenAICompletions: StreamFunction<"openai-completions", OpenA
const client = createClient(model, context, apiKey, options?.headers);
const params = buildParams(model, context, options);
options?.onPayload?.(params);
const openaiStream = await client.chat.completions.create(params, { signal: options?.signal });
const openaiStream = await client.chat.completions.create(params, {
signal: options?.signal,
});
stream.push({ type: "start", partial: output });

let currentBlock: TextContent | ThinkingContent | (ToolCall & { partialArgs?: string }) | null = null;
Expand Down Expand Up @@ -183,7 +185,11 @@ export const streamOpenAICompletions: StreamFunction<"openai-completions", OpenA
finishCurrentBlock(currentBlock);
currentBlock = { type: "text", text: "" };
output.content.push(currentBlock);
stream.push({ type: "text_start", contentIndex: blockIndex(), partial: output });
stream.push({
type: "text_start",
contentIndex: blockIndex(),
partial: output,
});
}

if (currentBlock.type === "text") {
Expand Down Expand Up @@ -225,7 +231,11 @@ export const streamOpenAICompletions: StreamFunction<"openai-completions", OpenA
thinkingSignature: foundReasoningField,
};
output.content.push(currentBlock);
stream.push({ type: "thinking_start", contentIndex: blockIndex(), partial: output });
stream.push({
type: "thinking_start",
contentIndex: blockIndex(),
partial: output,
});
}

if (currentBlock.type === "thinking") {
Expand Down Expand Up @@ -256,7 +266,11 @@ export const streamOpenAICompletions: StreamFunction<"openai-completions", OpenA
partialArgs: "",
};
output.content.push(currentBlock);
stream.push({ type: "toolcall_start", contentIndex: blockIndex(), partial: output });
stream.push({
type: "toolcall_start",
contentIndex: blockIndex(),
partial: output,
});
}

if (currentBlock.type === "toolCall") {
Expand Down Expand Up @@ -439,7 +453,9 @@ function buildParams(model: Model<"openai-completions">, context: Context, optio
if (compat.thinkingFormat === "zai" && model.reasoning) {
// Z.ai uses binary thinking: { type: "enabled" | "disabled" }
// Must explicitly disable since z.ai defaults to thinking enabled
(params as any).thinking = { type: options?.reasoningEffort ? "enabled" : "disabled" };
(params as any).thinking = {
type: options?.reasoningEffort ? "enabled" : "disabled",
};
} else if (options?.reasoningEffort && model.reasoning && compat.supportsReasoningEffort) {
// OpenAI-style reasoning_effort
params.reasoning_effort = options.reasoningEffort;
Expand Down Expand Up @@ -508,7 +524,10 @@ export function convertMessages(
if (context.systemPrompt) {
const useDeveloperRole = model.reasoning && compat.supportsDeveloperRole;
const role = useDeveloperRole ? "developer" : "system";
params.push({ role: role, content: sanitizeSurrogates(context.systemPrompt) });
params.push({
role: role,
content: sanitizeSurrogates(context.systemPrompt),
});
}

let lastRole: string | null = null;
Expand Down Expand Up @@ -585,7 +604,10 @@ export function convertMessages(
if (compat.requiresThinkingAsText) {
// Convert thinking blocks to plain text (no tags to avoid model mimicking them)
const thinkingText = nonEmptyThinkingBlocks.map((b) => b.thinking).join("\n\n");
const textContent = assistantMsg.content as Array<{ type: "text"; text: string }> | null;
const textContent = assistantMsg.content as Array<{
type: "text";
text: string;
}> | null;
if (textContent) {
textContent.unshift({ type: "text", text: thinkingText });
} else {
Expand Down
2 changes: 1 addition & 1 deletion packages/ai/src/types.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,7 +31,7 @@ export type KnownProvider =
| "cerebras"
| "openrouter"
| "vercel-ai-gateway"
| "zai"
| "zai-coding-plan"
| "mistral"
| "minimax"
| "minimax-cn"
Expand Down
2 changes: 1 addition & 1 deletion packages/ai/test/context-overflow.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -374,7 +374,7 @@

describe.skipIf(!process.env.ZAI_API_KEY)("z.ai", () => {
it("glm-4.5-flash - should detect overflow via isContextOverflow (silent overflow or rate limit)", async () => {
const model = getModel("zai", "glm-4.5-flash");
const model = getModel("zai-coding-plan", "glm-4.5-flash");
const result = await testContextOverflow(model, process.env.ZAI_API_KEY!);
logResult(result);

Expand Down Expand Up @@ -450,7 +450,7 @@
describe.skipIf(!process.env.OPENROUTER_API_KEY)("OpenRouter", () => {
// Anthropic backend
it("anthropic/claude-sonnet-4 via OpenRouter - should detect overflow via isContextOverflow", async () => {
const model = getModel("openrouter", "anthropic/claude-sonnet-4");

Check failure on line 453 in packages/ai/test/context-overflow.test.ts

View workflow job for this annotation

GitHub Actions / build-check-test 

Argument of type '"anthropic/claude-sonnet-4"' is not assignable to parameter of type '"openrouter/auto"'.
const result = await testContextOverflow(model, process.env.OPENROUTER_API_KEY!);
logResult(result);

Expand All @@ -461,7 +461,7 @@

// DeepSeek backend
it("deepseek/deepseek-v3.2 via OpenRouter - should detect overflow via isContextOverflow", async () => {
const model = getModel("openrouter", "deepseek/deepseek-v3.2");

Check failure on line 464 in packages/ai/test/context-overflow.test.ts

View workflow job for this annotation

GitHub Actions / build-check-test 

Argument of type '"deepseek/deepseek-v3.2"' is not assignable to parameter of type '"openrouter/auto"'.
const result = await testContextOverflow(model, process.env.OPENROUTER_API_KEY!);
logResult(result);

Expand All @@ -472,7 +472,7 @@

// Mistral backend
it("mistralai/mistral-large-2512 via OpenRouter - should detect overflow via isContextOverflow", async () => {
const model = getModel("openrouter", "mistralai/mistral-large-2512");

Check failure on line 475 in packages/ai/test/context-overflow.test.ts

View workflow job for this annotation

GitHub Actions / build-check-test 

Argument of type '"mistralai/mistral-large-2512"' is not assignable to parameter of type '"openrouter/auto"'.
const result = await testContextOverflow(model, process.env.OPENROUTER_API_KEY!);
logResult(result);

Expand All @@ -483,7 +483,7 @@

// Google backend
it("google/gemini-2.5-flash via OpenRouter - should detect overflow via isContextOverflow", async () => {
const model = getModel("openrouter", "google/gemini-2.5-flash");

Check failure on line 486 in packages/ai/test/context-overflow.test.ts

View workflow job for this annotation

GitHub Actions / build-check-test 

Argument of type '"google/gemini-2.5-flash"' is not assignable to parameter of type '"openrouter/auto"'.
const result = await testContextOverflow(model, process.env.OPENROUTER_API_KEY!);
logResult(result);

Expand All @@ -494,7 +494,7 @@

// Meta/Llama backend
it("meta-llama/llama-4-maverick via OpenRouter - should detect overflow via isContextOverflow", async () => {
const model = getModel("openrouter", "meta-llama/llama-4-maverick");

Check failure on line 497 in packages/ai/test/context-overflow.test.ts

View workflow job for this annotation

GitHub Actions / build-check-test 

Argument of type '"meta-llama/llama-4-maverick"' is not assignable to parameter of type '"openrouter/auto"'.
const result = await testContextOverflow(model, process.env.OPENROUTER_API_KEY!);
logResult(result);

Expand Down
2 changes: 1 addition & 1 deletion packages/ai/test/empty.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -309,7 +309,7 @@ describe("AI Providers Empty Message Tests", () => {
});

describe.skipIf(!process.env.ZAI_API_KEY)("zAI Provider Empty Messages", () => {
const llm = getModel("zai", "glm-4.5-air");
const llm = getModel("zai-coding-plan", "glm-4.5-air");

it("should handle empty content array", { retry: 3, timeout: 30000 }, async () => {
await testEmptyMessage(llm);
Expand Down
Loading
Loading