# Hostlinku Docs ## Dokumentasi ### Mulai - [Pengenalan](https://docs.hostlinku.com/index.md): Apa itu Hostlinku, cara kerja arsitekturnya, dan urutan membaca dokumentasi ini - [Memulai Cepat](https://docs.hostlinku.com/getting-started.md): Dari daftar akun sampai request pertama ke gateway Hostlinku dalam lima menit - [Autentikasi](https://docs.hostlinku.com/authentication.md): Cara memakai API key Hostlinku, masa berlakunya, limit kredit per API key, dan praktik keamanannya ### Akun - [Pemakaian & Limit](https://docs.hostlinku.com/usage-and-limits.md): Rate limit gateway, limit kredit per API key, dan cara memantau pemakaian di console - [Billing & Saldo](https://docs.hostlinku.com/billing.md): Cara kerja saldo kredit Hostlinku, metode top-up, dan dasar perhitungan biaya per token ### Bantuan - [Dukungan](https://docs.hostlinku.com/support.md): Cara menghubungi tim Hostlinku dan informasi yang perlu disertakan saat melapor ## FAQs - [FAQ](https://docs.hostlinku.com/faqs/index.md): Pertanyaan yang paling sering muncul tentang gateway Hostlinku - [Stream Mode](https://docs.hostlinku.com/faqs/stream-mode.md): Cara memakai streaming SSE di Hostlinku dengan OpenAI SDK, Python requests, curl, dan Node.js. - [Error yang sering muncul](https://docs.hostlinku.com/faqs/errors.md): Penyebab dan solusi kode error Hostlinku yang paling sering ditemui. - [Rate limit](https://docs.hostlinku.com/faqs/rate-limits.md): Batas request Hostlinku, header yang dikirim, dan cara melakukan backoff. - [Billing & top-up](https://docs.hostlinku.com/faqs/billing.md): Pertanyaan umum tentang saldo, top-up, limit kredit API key, dan error 402. - [Model tidak tersedia](https://docs.hostlinku.com/faqs/model-unavailable.md): Perbedaan 404 dan 503, Service Mode, model deprecated, dan langkah yang bisa dilakukan. ## API ### API - [API Reference](https://docs.hostlinku.com/api-reference.md): Referensi ringkas satu halaman - rate limit, Anthropic Messages API, OpenAI Chat Completions, OpenAI Responses API, kode error, dan daftar model - [SDK & Contoh Kode](https://docs.hostlinku.com/sdk.md): Contoh Python dan Node.js untuk tiga dialek API Hostlinku - Anthropic Messages, OpenAI Chat Completions, dan OpenAI Responses. - [Chat Completions](https://docs.hostlinku.com/chat-completions.md): Referensi endpoint POST /v1/chat/completions, mencakup parameter, streaming SSE, dan contoh curl + Python - [Anthropic Messages](https://docs.hostlinku.com/anthropic-messages.md): Referensi endpoint POST /v1/messages (Anthropic-compatible) - pakai Anthropic SDK, Claude Code, atau Cursor dengan API key Hostlinku - [Responses API](https://docs.hostlinku.com/responses.md): Referensi endpoint POST /v1/responses (OpenAI Responses API-compatible) - pakai Codex CLI, OpenAI Agents SDK, atau client.responses.create() dengan API key Hostlinku - [Hemat dengan Prompt Caching](https://docs.hostlinku.com/prompt-caching.md): Cara memakai cache_control di Hostlinku agar prompt besar yang berulang hanya ditagih ~10% - untuk Claude Code, Cursor, agent, dan aplikasi dengan system prompt panjang - [Model](https://docs.hostlinku.com/models.md): Endpoint GET /v1/models, bentuk data katalog, dan cara memilih model yang tepat - [Service Mode Dedicated](https://docs.hostlinku.com/dedicated.md): Model yang berjalan di infrastruktur Hostlinku sendiri - DeepSeek, GLM, dan Kimi dengan konteks 1 juta token. - [Error](https://docs.hostlinku.com/errors.md): Format envelope error Hostlinku, daftar kode error, dan cara menanganinya ### Model - [Qwen 3.6 Plus](https://docs.hostlinku.com/models/qwen3-6-plus.md): Qwen3.6-Plus is a hosted Alibaba Cloud Model Studio model for agentic coding, multimodal perception and reasoning, tool use, and real-world business task exe... - [Qwen 3.7 Max](https://docs.hostlinku.com/models/qwen3-7-max.md): Qwen3.7-Max is the flagship model in the Qwen3.7 series, built for agent workloads, long-chain reasoning, code understanding, productivity, and complex task... - [Model Baize](https://docs.hostlinku.com/models/model-baize.md): Model Model Baize dari Anonymous. - [Claude Fable 5](https://docs.hostlinku.com/models/claude-fable-5.md): Claude Fable 5 is Anthropic's flagship Mythos-class model optimized for long-horizon reasoning, agentic coding, and autonomous workflows. It supports text an... - [Claude Fable 5.1](https://docs.hostlinku.com/models/claude-fable-5-1.md): Model Claude Fable 5.1 dari Anthropic. - [Claude Haiku 4.5](https://docs.hostlinku.com/models/claude-haiku-4-5.md): Claude Haiku 4.5 is Anthropic's fastest near-frontier model, supporting 200K context, 64K max output, and extended thinking for low-latency, high-volume use. - [Claude Opus 4.6](https://docs.hostlinku.com/models/claude-opus-4-6.md): Claude Opus 4.6 is a high-capability Opus model for complex coding, professional knowledge work, enterprise workflows, and long-running agent tasks. - [Claude Opus 4.7](https://docs.hostlinku.com/models/claude-opus-4-7.md): Claude Opus 4.7 is a previous high-capability Opus model for complex coding, professional knowledge work, long-running agents, and difficult sustained tasks. - [Claude Opus 4.8](https://docs.hostlinku.com/models/claude-opus-4-8.md): Claude Opus 4.8 is Anthropic's high-capability Opus model for complex reasoning and agentic coding, with text/image input, text output, a 1M context window,... - [Claude Opus 5](https://docs.hostlinku.com/models/claude-opus-5.md): A high-capability model for complex agentic coding and enterprise work, with stronger deep reasoning, long-horizon execution, code review, office and documen... - [Claude Opus 5.5](https://docs.hostlinku.com/models/claude-opus-5-5.md): Anthropic's model for long-running agentic coding and knowledge work. Adaptive thinking always on (default effort medium), 1M context, 128K max output; cache... - [Claude Sonnet 4.6](https://docs.hostlinku.com/models/claude-sonnet-4-6.md): Claude Sonnet 4.6 balances speed and intelligence, with 1M context, 64K max output, and extended thinking for coding, agents, knowledge work, and long-contex... - [Claude Sonnet 5](https://docs.hostlinku.com/models/claude-sonnet-5.md): Claude Sonnet 5 is designed to be the most agent-capable Sonnet model to date. It can plan, use tools such as browsers and terminals, and operate autonomousl... - [Claude Sonnet 5.5](https://docs.hostlinku.com/models/claude-sonnet-5-5.md): Model seimbang untuk tugas sehari-hari. Kualitas, kecepatan, dan biaya yang pas. - [DeepSeek V4 Flash](https://docs.hostlinku.com/models/deepseek-v4-flash.md): DeepSeek V4 Flash is the efficiency-oriented V4 variant, supporting a 1M context window for everyday chat, summarization, code generation, structured process... - [DeepSeek V4 Pro](https://docs.hostlinku.com/models/deepseek-v4-pro.md): DeepSeek V4 Pro is the high-capability V4 variant, supporting a 1M context window for complex reasoning, agentic coding, long-document analysis, and high-qua... - [DeepSeek-V4.1-Flash](https://docs.hostlinku.com/models/deepseek-v4-1-flash.md): DeepSeek-V4.1-Flash is a multimodal mixture-of-experts model built on an asymmetric Causal Encoder–Decoder architecture, with native visual understanding. It... - [DeepSeek-V4.1-Flash Hyper](https://docs.hostlinku.com/models/deepseek-v4-1-flash-hyper.md): DeepSeek-V4.1-Flash is a multimodal mixture-of-experts model built on an asymmetric Causal Encoder–Decoder architecture, with native visual understanding. It... - [DeepSeek-V4.1-Flash Hyper Chat](https://docs.hostlinku.com/models/deepseek-v4-1-flash-hyper-chat.md): DeepSeek-V4.1-Flash is a multimodal mixture-of-experts model built on an asymmetric Causal Encoder–Decoder architecture, with native visual understanding. It... - [DeepSeek-V4.1-Flash Turbo](https://docs.hostlinku.com/models/deepseek-v4-1-flash-turbo.md): DeepSeek-V4.1-Flash is a multimodal mixture-of-experts model built on an asymmetric Causal Encoder–Decoder architecture, with native visual understanding. It... - [MiniMax M3](https://docs.hostlinku.com/models/minimax-m3.md): Latest M-series language models for agent inference, tool invocation, code and long-context tasks. - [Kimi K2.6](https://docs.hostlinku.com/models/kimi-k2-6.md): Kimi K2.6 is Kimi's latest and most intelligent model, featuring enhanced and more stable long-range coding capabilities and significantly improved instructi... - [Kimi K2.7 Code](https://docs.hostlinku.com/models/kimi-k2-7-code.md): Kimi K2.7 Code is Kimi’s most intelligent coding model to date. It follows instructions more reliably in long contexts and can complete programming tasks wit... - [Kimi K3](https://docs.hostlinku.com/models/kimi-k3.md): Kimi K3 is Kimi's flagship model, designed for long-range programming and end-to-end knowledge work; it features a 1-million-token context window and deliver... - [Kimi K3 Hyper](https://docs.hostlinku.com/models/kimi-k3-hyper.md): Kimi K3 is Kimi's flagship model, designed for long-range programming and end-to-end knowledge work; it features a 1-million-token context window and deliver... - [Kimi K3 Hyper Chat](https://docs.hostlinku.com/models/kimi-k3-hyper-chat.md): Kimi K3 is Kimi's flagship model, designed for long-range programming and end-to-end knowledge work; it features a 1-million-token context window and deliver... - [Kimi K3 Turbo](https://docs.hostlinku.com/models/kimi-k3-turbo.md): Kimi K3 is Kimi's flagship model, designed for long-range programming and end-to-end knowledge work; it features a 1-million-token context window and deliver... - [GPT 5.3 Codex](https://docs.hostlinku.com/models/gpt-5-3-codex.md): GPT-5.3-Codex is OpenAI's model for agentic coding, with text/image input, 400K context, 128K max output, function calling, and structured outputs. - [GPT-5.5](https://docs.hostlinku.com/models/gpt-5-5.md): GPT-5.5 is OpenAI's flagship model for complex professional work, with text/image input, text output, 1.05M context, 128K max output, and Responses tool supp... - [GPT-5.6 Luna](https://docs.hostlinku.com/models/gpt-5-6-luna.md): OpenAI's fastest and most cost-efficient model, designed for high-throughput chat, workflow automation, and large-scale token processing. - [GPT-5.6 Sol](https://docs.hostlinku.com/models/gpt-5-6-sol.md): OpenAI's most capable flagship model, built for complex reasoning, advanced coding, and intelligent agent workflows. - [GPT-5.6 Terra](https://docs.hostlinku.com/models/gpt-5-6-terra.md): The best balance of performance, speed, and cost, making it the ideal default model for businesses and developers. - [GPT-6 Astra](https://docs.hostlinku.com/models/gpt-6-astra.md): OpenAI's flagship model for complex reasoning, coding, and professional work, with text and image input, text output, a 1.05M context window, and 128K max ou... - [GPT-6 Luna](https://docs.hostlinku.com/models/gpt-6-luna.md): OpenAI's most efficient model for focused, high-volume tasks. Text and image input, text output; reasoning.effort none/low/medium/high/xhigh/max; 1,050,000 c... - [GPT-6 Sol](https://docs.hostlinku.com/models/gpt-6-sol.md): OpenAI's model built for complex coding and agentic workflows. Text and image input, text output; reasoning.effort none/low/medium/high/xhigh/max; 1,050,000... - [GPT-6.1 Sol](https://docs.hostlinku.com/models/gpt-6-1-sol.md): OpenAI's efficient frontier reasoning model in the GPT-6 Sol family for complex coding, computer use, and professional work. Text and image input, text outpu... - [MiMo V2.5 Pro](https://docs.hostlinku.com/models/mimo-v2-5-pro.md): The flagship base model with trillions of parameters can stably complete long-distance tasks involving nearly a thousand rounds of tool calls. Agent capabili... - [GLM 5](https://docs.hostlinku.com/models/glm-5.md): GLM-5 is Zhipu's text foundation model for agentic engineering, supporting 200K context, 128K max output, thinking modes, function calling, structured output... - [GLM 5.1](https://docs.hostlinku.com/models/glm-5-1.md): GLM-5.1 is Zhipu's flagship text model, with 200K context and 128K max output, strengthened for long-horizon planning, agentic coding, tool use, structured o... - [GLM 5.2](https://docs.hostlinku.com/models/glm-5-2.md): GLM-5.2 is a flagship model designed for the era of long-context tasks. It supports a truly usable 1-million-token context window-capable of handling project... - [GLM 5.3](https://docs.hostlinku.com/models/glm-5-3.md): GLM-5.3 is Zhipu AI's latest flagship text model for complex software engineering and long-horizon agent tasks. It improves coding, terminal operation, and l... - [GLM 5.3 Hyper Chat](https://docs.hostlinku.com/models/glm-5-3-hyper-chat.md): GLM-5.3 is Zhipu AI's latest flagship text model for complex software engineering and long-horizon agent tasks. It improves coding, terminal operation, and l... - [GLM-5.3 Flash](https://docs.hostlinku.com/models/glm-5-3-flash.md): GLM-5.3-Flash is Zhipu AI's lighter multimodal sibling to GLM-5.3 for everyday analysis and agentic work. Available on the Hostlinku Dedicated channel. - [GLM-5.3 Flash Hyper](https://docs.hostlinku.com/models/glm-5-3-flash-hyper.md): GLM-5.3-Flash is Zhipu AI's lighter multimodal sibling to GLM-5.3 for everyday analysis and agentic work. Available on the Hostlinku Dedicated channel with a... - [GLM-5.3 Flash Hyper Chat](https://docs.hostlinku.com/models/glm-5-3-flash-hyper-chat.md): GLM-5.3-Flash is Zhipu AI's lighter multimodal sibling to GLM-5.3 for everyday analysis and agentic work. Available on the Hostlinku Dedicated channel with a... ## Integrasi ### Integrasi - [Ikhtisar Integrasi](https://docs.hostlinku.com/integrations/index.md): Pakai Hostlinku di Claude Code, Codex CLI, Cursor, VS Code, GitHub Copilot, Zed, Cline, Kilo Code, Continue, JetBrains AI Assistant, Junie, Factory Droid, OpenCode, Xum, Aider, dan OpenAI Agents SDK - cukup ganti base URL, tanpa mengubah cara kerja tool #### Integrasi IDE & Agent - [Claude Code](https://docs.hostlinku.com/integrations/claude-code.md): Jalankan Claude Code (CLI, VS Code, JetBrains) lewat Hostlinku dengan ANTHROPIC_BASE_URL - pakai Claude atau model lain, satu saldo untuk semuanya - [Codex CLI](https://docs.hostlinku.com/integrations/codex-cli.md): Jalankan OpenAI Codex CLI lewat Hostlinku dengan provider kustom di ~/.codex/config.toml (wire_api = "responses") - [Cursor](https://docs.hostlinku.com/integrations/cursor.md): Pakai model Hostlinku di Cursor lewat Override OpenAI Base URL - chat, Cmd-K, dan Agent dengan API key kamu sendiri - [VS Code (Chat)](https://docs.hostlinku.com/integrations/vscode.md): Pakai model Hostlinku di chat & agent bawaan VS Code lewat provider Custom Endpoint - tanpa login GitHub, cukup API key Hostlinku - [GitHub Copilot](https://docs.hostlinku.com/integrations/copilot.md): Pakai model Hostlinku di GitHub Copilot CLI dan GitHub Copilot app lewat BYOK, plus batasan Copilot di IDE - [Zed](https://docs.hostlinku.com/integrations/zed.md): Tambahkan Hostlinku sebagai provider openai_compatible di settings.json Zed dan pakai model apa pun di Agent Panel - [Cline & Roo Code](https://docs.hostlinku.com/integrations/cline.md): Konfigurasi provider OpenAI Compatible (atau Anthropic dengan base URL kustom) di Cline dan Roo Code untuk memakai model Hostlinku - [Kilo Code](https://docs.hostlinku.com/integrations/kilo-code.md): Buat Configuration Profile Hostlinku di Kilo Code lewat provider OpenAI (custom base URL), OpenAI Compatible, atau Anthropic - [Continue](https://docs.hostlinku.com/integrations/continue.md): Tambahkan model Hostlinku ke config.yaml Continue (VS Code & JetBrains) dengan provider openai dan apiBase kustom - [JetBrains AI Assistant](https://docs.hostlinku.com/integrations/jetbrains.md): Pakai model Hostlinku di IntelliJ IDEA, PyCharm, WebStorm, GoLand, dan IDE JetBrains lain lewat provider OpenAI-compatible di AI Assistant - [Junie CLI](https://docs.hostlinku.com/integrations/junie.md): Hubungkan Junie CLI dari JetBrains ke Hostlinku lewat custom model profile (Chat Completions atau Anthropic Messages) - [Factory Droid](https://docs.hostlinku.com/integrations/factory.md): Tambahkan model Hostlinku sebagai customModels di ~/.factory/settings.json dan pakai di Droid CLI maupun aplikasi desktop Factory - [OpenCode](https://docs.hostlinku.com/integrations/opencode.md): Daftarkan Hostlinku sebagai provider kustom OpenAI-compatible di opencode.json dan pilih model lewat /models - [Xum](https://docs.hostlinku.com/integrations/xum.md): Daftarkan Hostlinku sebagai custom provider di Xum (dulu Mux) untuk menjalankan banyak coding agent paralel dengan satu API key - [Aider](https://docs.hostlinku.com/integrations/aider.md): Jalankan Aider lewat Hostlinku dengan OPENAI_API_BASE dan prefix model openai/ - plus metadata model agar konteks dan biaya dihitung benar - [OpenAI Agents SDK](https://docs.hostlinku.com/integrations/agents-sdk.md): Pakai OpenAI Agents SDK (Python & TypeScript) dengan Hostlinku lewat Responses API - set_default_openai_client, tracing dimatikan, tools & handoff tetap berjalan ## Legal ### Dokumen Legal - [Syarat dan Ketentuan](https://docs.hostlinku.com/legals/terms-of-service.md): Syarat dan Ketentuan penggunaan Hostlinku: akun, API key, saldo prabayar, penagihan per request, batasan layanan, tanggung jawab, dan penyelesaian sengketa. - [Kebijakan Pengembalian Dana](https://docs.hostlinku.com/legals/refund-policy.md): Kebijakan pengembalian dana Hostlinku: pelepasan reservasi saldo otomatis, refund sisa saldo top-up, top-up gagal, tata cara pengajuan, dan pengecualian. - [Kebijakan Privasi](https://docs.hostlinku.com/legals/privacy-policy.md): Kebijakan privasi Hostlinku: data akun, API key, metadata request, body logging, pembayaran, pihak penerima data, retensi, keamanan, dan hak subjek data sesuai UU PDP. - [Kebijakan Penggunaan](https://docs.hostlinku.com/legals/acceptable-use-policy.md): Kebijakan penggunaan API Hostlinku: penggunaan yang diizinkan, konten dan aktivitas terlarang, rate limit, penegakan, dan cara melaporkan pelanggaran. - [Keamanan](https://docs.hostlinku.com/legals/security.md): Praktik keamanan Hostlinku: enkripsi, autentikasi, penanganan API key, isolasi data, audit log, respons insiden, dan cara melaporkan kerentanan. > The links below point to documentation indexes. Follow each `/_llms/` index recursively until you reach documentation pages. ## Indexes - [English (88 pages)](https://docs.hostlinku.com/_llms/en.txt): Documentation for English. - [Japanese (88 pages)](https://docs.hostlinku.com/_llms/ja.txt): Documentation for Japanese. - [Chinese (88 pages)](https://docs.hostlinku.com/_llms/zh.txt): Documentation for Chinese. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.