diff --git a/packages/ai/CHANGELOG.md b/packages/ai/CHANGELOG.md index f9c1f39e..5c329353 100644 --- a/packages/ai/CHANGELOG.md +++ b/packages/ai/CHANGELOG.md @@ -2,6 +2,10 @@ ## [Unreleased] +### Added + +- Added Kimi For Coding provider support (Moonshot AI's Anthropic-compatible coding API) + ## [0.50.2] - 2026-01-29 ### Added diff --git a/packages/ai/README.md b/packages/ai/README.md index 6207e4b7..8f5c1b79 100644 --- a/packages/ai/README.md +++ b/packages/ai/README.md @@ -63,6 +63,7 @@ Unified LLM API with automatic model discovery, provider configuration, token an - **Google Gemini CLI** (requires OAuth, see below) - **Antigravity** (requires OAuth, see below) - **Amazon Bedrock** +- **Kimi For Coding** (Moonshot AI, uses Anthropic-compatible API) - **Any OpenAI-compatible API**: Ollama, vLLM, LM Studio, etc. ## Installation @@ -894,6 +895,7 @@ In Node.js environments, you can set environment variables to avoid passing API | Vercel AI Gateway | `AI_GATEWAY_API_KEY` | | zAI | `ZAI_API_KEY` | | MiniMax | `MINIMAX_API_KEY` | +| Kimi For Coding | `KIMI_API_KEY` | | GitHub Copilot | `COPILOT_GITHUB_TOKEN` or `GH_TOKEN` or `GITHUB_TOKEN` | When set, the library automatically uses these keys: diff --git a/packages/ai/scripts/generate-models.ts b/packages/ai/scripts/generate-models.ts index bc97b3f4..73e077d2 100644 --- a/packages/ai/scripts/generate-models.ts +++ b/packages/ai/scripts/generate-models.ts @@ -630,6 +630,33 @@ async function loadModelsDevData(): Promise[]> { } } + // Process Kimi For Coding models + if (data["kimi-for-coding"]?.models) { + for (const [modelId, model] of Object.entries(data["kimi-for-coding"].models)) { + const m = model as ModelsDevModel; + if (m.tool_call !== true) continue; + + models.push({ + id: modelId, + name: m.name || modelId, + api: "anthropic-messages", + provider: "kimi-coding", + // Kimi For Coding's Anthropic-compatible API - SDK appends /v1/messages + baseUrl: "https://api.kimi.com/coding", + reasoning: m.reasoning === true, + input: m.modalities?.input?.includes("image") ? ["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, + }, + contextWindow: m.limit?.context || 4096, + maxTokens: m.limit?.output || 4096, + }); + } + } + console.log(`Loaded ${models.length} tool-capable models from models.dev`); return models; } catch (error) { @@ -1130,6 +1157,42 @@ async function generateModels() { ]; allModels.push(...vertexModels); + // Kimi For Coding models (Moonshot AI's Anthropic-compatible coding API) + // Static fallback in case models.dev doesn't have them yet + const KIMI_CODING_BASE_URL = "https://api.kimi.com/coding"; + const kimiCodingModels: Model<"anthropic-messages">[] = [ + { + id: "kimi-k2-thinking", + name: "Kimi K2 Thinking", + api: "anthropic-messages", + provider: "kimi-coding", + baseUrl: KIMI_CODING_BASE_URL, + reasoning: true, + input: ["text"], + cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 }, + contextWindow: 262144, + maxTokens: 32768, + }, + { + id: "k2p5", + name: "Kimi K2.5", + api: "anthropic-messages", + provider: "kimi-coding", + baseUrl: KIMI_CODING_BASE_URL, + reasoning: true, + input: ["text"], + cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 }, + contextWindow: 262144, + maxTokens: 32768, + }, + ]; + // Only add if not already present from models.dev + for (const model of kimiCodingModels) { + if (!allModels.some(m => m.provider === "kimi-coding" && m.id === model.id)) { + allModels.push(model); + } + } + const azureOpenAiModels: Model[] = allModels .filter((model) => model.provider === "openai" && model.api === "openai-responses") .map((model) => ({ diff --git a/packages/ai/src/env-api-keys.ts b/packages/ai/src/env-api-keys.ts index e82e9f0c..fe759320 100644 --- a/packages/ai/src/env-api-keys.ts +++ b/packages/ai/src/env-api-keys.ts @@ -107,6 +107,7 @@ export function getEnvApiKey(provider: any): string | undefined { "minimax-cn": "MINIMAX_CN_API_KEY", huggingface: "HF_TOKEN", opencode: "OPENCODE_API_KEY", + "kimi-coding": "KIMI_API_KEY", }; const envVar = envMap[provider]; diff --git a/packages/ai/src/models.generated.ts b/packages/ai/src/models.generated.ts index 57f23166..81429734 100644 --- a/packages/ai/src/models.generated.ts +++ b/packages/ai/src/models.generated.ts @@ -3536,6 +3536,42 @@ export const MODELS = { maxTokens: 128000, } satisfies Model<"openai-completions">, }, + "kimi-coding": { + "k2p5": { + id: "k2p5", + name: "Kimi K2.5", + api: "anthropic-messages", + provider: "kimi-coding", + baseUrl: "https://api.kimi.com/coding", + reasoning: true, + input: ["text", "image"], + cost: { + input: 0, + output: 0, + cacheRead: 0, + cacheWrite: 0, + }, + contextWindow: 262144, + maxTokens: 32768, + } satisfies Model<"anthropic-messages">, + "kimi-k2-thinking": { + id: "kimi-k2-thinking", + name: "Kimi K2 Thinking", + api: "anthropic-messages", + provider: "kimi-coding", + baseUrl: "https://api.kimi.com/coding", + reasoning: true, + input: ["text"], + cost: { + input: 0, + output: 0, + cacheRead: 0, + cacheWrite: 0, + }, + contextWindow: 262144, + maxTokens: 32768, + } satisfies Model<"anthropic-messages">, + }, "minimax": { "MiniMax-M2": { id: "MiniMax-M2", diff --git a/packages/ai/src/types.ts b/packages/ai/src/types.ts index 0da018aa..5d8eb266 100644 --- a/packages/ai/src/types.ts +++ b/packages/ai/src/types.ts @@ -36,7 +36,8 @@ export type KnownProvider = | "minimax" | "minimax-cn" | "huggingface" - | "opencode"; + | "opencode" + | "kimi-coding"; export type Provider = KnownProvider | string; export type ThinkingLevel = "minimal" | "low" | "medium" | "high" | "xhigh"; diff --git a/packages/ai/src/utils/overflow.ts b/packages/ai/src/utils/overflow.ts index 03d888b1..507a9a7c 100644 --- a/packages/ai/src/utils/overflow.ts +++ b/packages/ai/src/utils/overflow.ts @@ -18,6 +18,7 @@ import type { AssistantMessage } from "../types.js"; * - LM Studio: "tokens to keep from the initial prompt is greater than the context length" * - GitHub Copilot: "prompt token count of X exceeds the limit of Y" * - MiniMax: "invalid params, context window exceeds limit" + * - Kimi For Coding: "Your request exceeded model token limit: X (requested: Y)" * - Cerebras: Returns "400/413 status code (no body)" - handled separately below * - Mistral: Returns "400/413 status code (no body)" - handled separately below * - z.ai: Does NOT error, accepts overflow silently - handled via usage.input > contextWindow @@ -35,6 +36,7 @@ const OVERFLOW_PATTERNS = [ /exceeds the available context size/i, // llama.cpp server /greater than the context length/i, // LM Studio /context window exceeds limit/i, // MiniMax + /exceeded model token limit/i, // Kimi For Coding /context[_ ]length[_ ]exceeded/i, // Generic fallback /too many tokens/i, // Generic fallback /token limit exceeded/i, // Generic fallback @@ -62,6 +64,7 @@ const OVERFLOW_PATTERNS = [ * - OpenRouter (all backends): "maximum context length is X tokens" * - llama.cpp: "exceeds the available context size" * - LM Studio: "greater than the context length" + * - Kimi For Coding: "exceeded model token limit: X (requested: Y)" * * **Unreliable detection:** * - z.ai: Sometimes accepts overflow silently (detectable via usage.input > contextWindow), diff --git a/packages/ai/test/abort.test.ts b/packages/ai/test/abort.test.ts index 45a4e0f2..3de3fb04 100644 --- a/packages/ai/test/abort.test.ts +++ b/packages/ai/test/abort.test.ts @@ -193,6 +193,18 @@ describe("AI Providers Abort Tests", () => { }); }); + describe.skipIf(!process.env.KIMI_API_KEY)("Kimi For Coding Provider Abort", () => { + const llm = getModel("kimi-coding", "kimi-k2-thinking"); + + it("should abort mid-stream", { retry: 3 }, async () => { + await testAbortSignal(llm); + }); + + it("should handle immediate abort", { retry: 3 }, async () => { + await testImmediateAbort(llm); + }); + }); + describe.skipIf(!process.env.AI_GATEWAY_API_KEY)("Vercel AI Gateway Provider Abort", () => { const llm = getModel("vercel-ai-gateway", "google/gemini-2.5-flash"); diff --git a/packages/ai/test/context-overflow.test.ts b/packages/ai/test/context-overflow.test.ts index 98f2087a..8cb00486 100644 --- a/packages/ai/test/context-overflow.test.ts +++ b/packages/ai/test/context-overflow.test.ts @@ -443,6 +443,21 @@ describe("Context overflow error handling", () => { }, 120000); }); + // ============================================================================= + // Kimi For Coding + // ============================================================================= + + describe.skipIf(!process.env.KIMI_API_KEY)("Kimi For Coding", () => { + it("kimi-k2-thinking - should detect overflow via isContextOverflow", async () => { + const model = getModel("kimi-coding", "kimi-k2-thinking"); + const result = await testContextOverflow(model, process.env.KIMI_API_KEY!); + logResult(result); + + expect(result.stopReason).toBe("error"); + expect(isContextOverflow(result.response, model.contextWindow)).toBe(true); + }, 120000); + }); + // ============================================================================= // Vercel AI Gateway - Unified API for multiple providers // ============================================================================= diff --git a/packages/ai/test/cross-provider-handoff.test.ts b/packages/ai/test/cross-provider-handoff.test.ts index a9058478..2f5e0cb4 100644 --- a/packages/ai/test/cross-provider-handoff.test.ts +++ b/packages/ai/test/cross-provider-handoff.test.ts @@ -88,6 +88,8 @@ const PROVIDER_MODEL_PAIRS: ProviderModelPair[] = [ { provider: "groq", model: "openai/gpt-oss-120b", label: "groq-gpt-oss-120b" }, // Hugging Face { provider: "huggingface", model: "moonshotai/Kimi-K2.5", label: "huggingface-kimi-k2.5" }, + // Kimi For Coding + { provider: "kimi-coding", model: "kimi-k2-thinking", label: "kimi-coding-k2-thinking" }, // Mistral { provider: "mistral", model: "devstral-medium-latest", label: "mistral-devstral-medium" }, // MiniMax diff --git a/packages/ai/test/empty.test.ts b/packages/ai/test/empty.test.ts index e18d8d03..7231e1f1 100644 --- a/packages/ai/test/empty.test.ts +++ b/packages/ai/test/empty.test.ts @@ -388,6 +388,26 @@ describe("AI Providers Empty Message Tests", () => { }); }); + describe.skipIf(!process.env.KIMI_API_KEY)("Kimi For Coding Provider Empty Messages", () => { + const llm = getModel("kimi-coding", "kimi-k2-thinking"); + + it("should handle empty content array", { retry: 3, timeout: 30000 }, async () => { + await testEmptyMessage(llm); + }); + + it("should handle empty string content", { retry: 3, timeout: 30000 }, async () => { + await testEmptyStringMessage(llm); + }); + + it("should handle whitespace-only content", { retry: 3, timeout: 30000 }, async () => { + await testWhitespaceOnlyMessage(llm); + }); + + it("should handle empty assistant message in conversation", { retry: 3, timeout: 30000 }, async () => { + await testEmptyAssistantMessage(llm); + }); + }); + describe.skipIf(!process.env.AI_GATEWAY_API_KEY)("Vercel AI Gateway Provider Empty Messages", () => { const llm = getModel("vercel-ai-gateway", "google/gemini-2.5-flash"); diff --git a/packages/ai/test/image-tool-result.test.ts b/packages/ai/test/image-tool-result.test.ts index 11c6e562..124674f9 100644 --- a/packages/ai/test/image-tool-result.test.ts +++ b/packages/ai/test/image-tool-result.test.ts @@ -300,6 +300,18 @@ describe("Tool Results with Images", () => { }); }); + describe.skipIf(!process.env.KIMI_API_KEY)("Kimi For Coding Provider (k2p5)", () => { + const llm = getModel("kimi-coding", "k2p5"); + + it("should handle tool result with only image", { retry: 3, timeout: 30000 }, async () => { + await handleToolWithImageResult(llm); + }); + + it("should handle tool result with text and image", { retry: 3, timeout: 30000 }, async () => { + await handleToolWithTextAndImageResult(llm); + }); + }); + describe.skipIf(!process.env.AI_GATEWAY_API_KEY)("Vercel AI Gateway Provider (google/gemini-2.5-flash)", () => { const llm = getModel("vercel-ai-gateway", "google/gemini-2.5-flash"); diff --git a/packages/ai/test/stream.test.ts b/packages/ai/test/stream.test.ts index cef2f9bc..6b70550f 100644 --- a/packages/ai/test/stream.test.ts +++ b/packages/ai/test/stream.test.ts @@ -862,6 +862,33 @@ describe("Generate E2E Tests", () => { }); }); + describe.skipIf(!process.env.KIMI_API_KEY)( + "Kimi For Coding Provider (kimi-k2-thinking via Anthropic Messages)", + () => { + const llm = getModel("kimi-coding", "kimi-k2-thinking"); + + it("should complete basic text generation", { retry: 3 }, async () => { + await basicTextGeneration(llm); + }); + + it("should handle tool calling", { retry: 3 }, async () => { + await handleToolCall(llm); + }); + + it("should handle streaming", { retry: 3 }, async () => { + await handleStreaming(llm); + }); + + it("should handle thinking mode", { retry: 3 }, async () => { + await handleThinking(llm, { thinkingEnabled: true, thinkingBudgetTokens: 2048 }); + }); + + it("should handle multi-turn with thinking and tools", { retry: 3 }, async () => { + await multiTurn(llm, { thinkingEnabled: true, thinkingBudgetTokens: 2048 }); + }); + }, + ); + // ========================================================================= // OAuth-based providers (credentials from ~/.pi/agent/oauth.json) // Tokens are resolved at module level (see oauthTokens above) diff --git a/packages/ai/test/tokens.test.ts b/packages/ai/test/tokens.test.ts index c4709c47..65106fca 100644 --- a/packages/ai/test/tokens.test.ts +++ b/packages/ai/test/tokens.test.ts @@ -186,6 +186,14 @@ describe("Token Statistics on Abort", () => { }); }); + describe.skipIf(!process.env.KIMI_API_KEY)("Kimi For Coding Provider", () => { + const llm = getModel("kimi-coding", "kimi-k2-thinking"); + + it("should include token stats when aborted mid-stream", { retry: 3, timeout: 30000 }, async () => { + await testTokensOnAbort(llm); + }); + }); + describe.skipIf(!process.env.AI_GATEWAY_API_KEY)("Vercel AI Gateway Provider", () => { const llm = getModel("vercel-ai-gateway", "google/gemini-2.5-flash"); diff --git a/packages/ai/test/tool-call-without-result.test.ts b/packages/ai/test/tool-call-without-result.test.ts index 8caf22d0..9ad2b9fa 100644 --- a/packages/ai/test/tool-call-without-result.test.ts +++ b/packages/ai/test/tool-call-without-result.test.ts @@ -200,6 +200,14 @@ describe("Tool Call Without Result Tests", () => { }); }); + describe.skipIf(!process.env.KIMI_API_KEY)("Kimi For Coding Provider", () => { + const model = getModel("kimi-coding", "kimi-k2-thinking"); + + it("should filter out tool calls without corresponding tool results", { retry: 3, timeout: 30000 }, async () => { + await testToolCallWithoutResult(model); + }); + }); + describe.skipIf(!process.env.AI_GATEWAY_API_KEY)("Vercel AI Gateway Provider", () => { const model = getModel("vercel-ai-gateway", "google/gemini-2.5-flash"); diff --git a/packages/ai/test/total-tokens.test.ts b/packages/ai/test/total-tokens.test.ts index b5ce961a..6b1ff237 100644 --- a/packages/ai/test/total-tokens.test.ts +++ b/packages/ai/test/total-tokens.test.ts @@ -394,6 +394,29 @@ describe("totalTokens field", () => { ); }); + // ========================================================================= + // Kimi For Coding + // ========================================================================= + + describe.skipIf(!process.env.KIMI_API_KEY)("Kimi For Coding", () => { + it( + "kimi-k2-thinking - should return totalTokens equal to sum of components", + { retry: 3, timeout: 60000 }, + async () => { + const llm = getModel("kimi-coding", "kimi-k2-thinking"); + + console.log(`\nKimi For Coding / ${llm.id}:`); + const { first, second } = await testTotalTokensWithCache(llm, { apiKey: process.env.KIMI_API_KEY }); + + logUsage("First request", first); + logUsage("Second request", second); + + assertTotalTokensEqualsComponents(first); + assertTotalTokensEqualsComponents(second); + }, + ); + }); + // ========================================================================= // Vercel AI Gateway // ========================================================================= diff --git a/packages/ai/test/unicode-surrogate.test.ts b/packages/ai/test/unicode-surrogate.test.ts index 38e5426e..8f9d4f0d 100644 --- a/packages/ai/test/unicode-surrogate.test.ts +++ b/packages/ai/test/unicode-surrogate.test.ts @@ -675,6 +675,22 @@ describe("AI Providers Unicode Surrogate Pair Tests", () => { }); }); + describe.skipIf(!process.env.KIMI_API_KEY)("Kimi For Coding Provider Unicode Handling", () => { + const llm = getModel("kimi-coding", "kimi-k2-thinking"); + + it("should handle emoji in tool results", { retry: 3, timeout: 30000 }, async () => { + await testEmojiInToolResults(llm); + }); + + it("should handle real-world LinkedIn comment data with emoji", { retry: 3, timeout: 30000 }, async () => { + await testRealWorldLinkedInData(llm); + }); + + it("should handle unpaired high surrogate (0xD83D) in tool results", { retry: 3, timeout: 30000 }, async () => { + await testUnpairedHighSurrogate(llm); + }); + }); + describe.skipIf(!process.env.AI_GATEWAY_API_KEY)("Vercel AI Gateway Provider Unicode Handling", () => { const llm = getModel("vercel-ai-gateway", "google/gemini-2.5-flash"); diff --git a/packages/coding-agent/CHANGELOG.md b/packages/coding-agent/CHANGELOG.md index d0cc3ff2..d69e84f5 100644 --- a/packages/coding-agent/CHANGELOG.md +++ b/packages/coding-agent/CHANGELOG.md @@ -2,6 +2,10 @@ ## [Unreleased] +### Added + +- Added Kimi For Coding provider support (Moonshot AI's Anthropic-compatible coding API). Set `KIMI_API_KEY` environment variable. See [README.md#kimi-for-coding](README.md#kimi-for-coding). + ## [0.50.2] - 2026-01-29 ### New Features diff --git a/packages/coding-agent/README.md b/packages/coding-agent/README.md index 5d0c0388..20c0e1ab 100644 --- a/packages/coding-agent/README.md +++ b/packages/coding-agent/README.md @@ -94,6 +94,7 @@ For each built-in provider, pi maintains a list of tool-capable models, updated - ZAI - OpenCode Zen - Hugging Face +- Kimi For Coding - MiniMax See [docs/providers.md](docs/providers.md) for detailed setup instructions. diff --git a/packages/coding-agent/docs/providers.md b/packages/coding-agent/docs/providers.md index 842a2ab6..f3ff0eea 100644 --- a/packages/coding-agent/docs/providers.md +++ b/packages/coding-agent/docs/providers.md @@ -63,6 +63,7 @@ pi | ZAI | `ZAI_API_KEY` | | OpenCode Zen | `OPENCODE_API_KEY` | | Hugging Face | `HF_TOKEN` | +| Kimi For Coding | `KIMI_API_KEY` | | MiniMax | `MINIMAX_API_KEY` | | MiniMax (China) | `MINIMAX_CN_API_KEY` | diff --git a/packages/coding-agent/src/cli/args.ts b/packages/coding-agent/src/cli/args.ts index fd108206..181731bf 100644 --- a/packages/coding-agent/src/cli/args.ts +++ b/packages/coding-agent/src/cli/args.ts @@ -280,6 +280,7 @@ ${chalk.bold("Environment Variables:")} ZAI_API_KEY - ZAI API key MISTRAL_API_KEY - Mistral API key MINIMAX_API_KEY - MiniMax API key + KIMI_API_KEY - Kimi For Coding API key AWS_PROFILE - AWS profile for Amazon Bedrock AWS_ACCESS_KEY_ID - AWS access key for Amazon Bedrock AWS_SECRET_ACCESS_KEY - AWS secret key for Amazon Bedrock diff --git a/packages/coding-agent/src/core/model-resolver.ts b/packages/coding-agent/src/core/model-resolver.ts index 8a7082b7..92fca0a8 100644 --- a/packages/coding-agent/src/core/model-resolver.ts +++ b/packages/coding-agent/src/core/model-resolver.ts @@ -33,6 +33,7 @@ export const defaultModelPerProvider: Record = { "minimax-cn": "MiniMax-M2.1", huggingface: "moonshotai/Kimi-K2.5", opencode: "claude-opus-4-5", + "kimi-coding": "kimi-k2-thinking", }; export interface ScopedModel {