Official sources · Basic access without login or Credits
Basic checks and downloads require no login or Credits. Getting a tool does not activate a subscription or configure an account.
Market?
Market labels help you discover tools; official regional support and access conditions are on each tool's detail. This filter is a browsing preference and never rewrites your actual location or account conditions.
Platform
Type
Purpose
89 tools to explore
Cursor
Desktop
International
Mainland China support unverified
AI desktop code editor.
Coding
Source checked: 2026-10-02 Source verification does not certify the installer.
Model-dependent monthly usage: $15–$60; 5-hour allowance: 20%, weekly: 50%. Request counts are estimates, not guaranteed; model allowances are not additive.
Model-dependent monthly usage: $15–$60; 5-hour allowance: 20%, weekly: 50%. Request counts are estimates, not guaranteed; model allowances are not additive.
Full seats include Pro-equivalent daily/weekly quota and Desktop. Flex seats include no quota or Desktop; all usage draws from shared on-demand credits.
GPT-4o · GPT-4.1 · o3 · o3 High Reasoning · Claude Opus 4.5 · Claude Opus 4.5 Thinking · Gemini 2.5 Pro · Gemini 3 Flash High · Gemini 3 Flash Low · Gemini 3 Flash Medium · Gemini 3 Flash Minimal · GPT-5.2 High Thinking · GPT-5.2 High Thinking Fast · GPT-5.2 Low Thinking · GPT-5.2 Low Thinking Fast · GPT-5.2 Medium Thinking · GPT-5.2 Medium Thinking Fast · GPT-5.2 No Thinking · GPT-5.2 No Thinking Fast · GPT-5.2 XHigh Thinking · GPT-5.2 XHigh Thinking Fast · GPT-OSS 120B Medium Thinking · Claude Haiku 4.5 · GPT-5.1 No Thinking · GPT-5.1 Low Thinking · GPT-5.1 Medium Thinking · GPT-5.1 High Thinking · GPT-5.1 No Thinking Fast · GPT-5.1 Low Thinking Fast · GPT-5.1 Medium Thinking Fast · GPT-5.1 High Thinking Fast · Grok Code Fast 1 · GPT-5 Low Thinking · GPT-5 Medium Thinking · GPT-5 High Thinking · Adaptive · Fast Arena · Hybrid Arena · Frontier Arena · Claude Fable 5 High · Claude Fable 5 Low · Claude Fable 5 Max · Claude Fable 5 Medium · Claude Fable 5 XHigh · Claude Fable 5.1 High · Claude Fable 5.1 Low · Claude Fable 5.1 Max · Claude Fable 5.1 Medium · Claude Fable 5.1 XHigh · Claude Opus 4.6 · Claude Opus 4.6 1M · Claude Opus 4.6 Thinking · Claude Opus 4.6 Thinking 1M · Claude Opus 4.7 High · Claude Opus 4.7 Low · Claude Opus 4.7 Max · Claude Opus 4.7 Medium · Claude Opus 4.7 XHigh · Claude Opus 4.8 High · Claude Opus 4.8 High Fast · Claude Opus 4.8 Low · Claude Opus 4.8 Low Fast · Claude Opus 4.8 Max · Claude Opus 4.8 Max Fast · Claude Opus 4.8 Medium · Claude Opus 4.8 Medium Fast · Claude Opus 4.8 XHigh · Claude Opus 4.8 XHigh Fast · Claude Opus 5 High · Claude Opus 5 High Fast · Claude Opus 5 Low · Claude Opus 5 Low Fast · Claude Opus 5 Max · Claude Opus 5 Max Fast · Claude Opus 5 Medium · Claude Opus 5 Medium Fast · Claude Opus 5 XHigh · Claude Opus 5 XHigh Fast · Claude Opus 5.5 High · Claude Opus 5.5 High Fast · Claude Opus 5.5 Low · Claude Opus 5.5 Low Fast · Claude Opus 5.5 Max · Claude Opus 5.5 Max Fast · Claude Opus 5.5 Medium · Claude Opus 5.5 Medium Fast · Claude Opus 5.5 XHigh · Claude Opus 5.5 XHigh Fast · Claude Sonnet 4.6 · Claude Sonnet 4.6 1M · Claude Sonnet 4.6 Thinking · Claude Sonnet 4.6 Thinking 1M · Claude Sonnet 5 High · Claude Sonnet 5 Low · Claude Sonnet 5 Max · Claude Sonnet 5 Medium · Claude Sonnet 5 XHigh · Claude Sonnet 5.5 High · Claude Sonnet 5.5 Low · Claude Sonnet 5.5 Max · Claude Sonnet 5.5 Medium · Claude Sonnet 5.5 XHigh · DeepSeek V4 Pro High · Gemini 3.1 Pro High Thinking · Gemini 3.1 Pro Low Thinking · Gemini 3.5 Flash High · Gemini 3.5 Flash Low · Gemini 3.5 Flash Medium · Gemini 3.5 Flash Minimal · Gemini 3.6 Flash High · Gemini 3.6 Flash Low · Gemini 3.6 Flash Medium · Gemini 3.6 Flash Minimal · Gemini 3.7 Flash High · Gemini 3.7 Flash Low · Gemini 3.7 Flash Medium · Gemini 3.8 Flash High · Gemini 3.8 Flash Low · Gemini 3.8 Flash Medium · GLM-5.2 High · GLM-5.2 High 1M · GLM-5.2 Max · GLM-5.2 Max 1M · GLM-5.2 No Thinking · GLM-5.2 No Thinking 1M · GPT-5.3-Codex High · GPT-5.3-Codex High Fast · GPT-5.3-Codex Low · GPT-5.3-Codex Low Fast · GPT-5.3-Codex Medium · GPT-5.3-Codex Medium Fast · GPT-5.3-Codex Spark Medium · GPT-5.3-Codex X-High · GPT-5.3-Codex XHigh Fast · GPT-5.4 High Thinking · GPT-5.4 High Thinking Fast · GPT-5.4 Low Thinking · GPT-5.4 Low Thinking Fast · GPT-5.4 Medium Thinking · GPT-5.4 Medium Thinking Fast · GPT-5.4 Mini High Thinking · GPT-5.4 Mini Low Thinking · GPT-5.4 Mini Medium Thinking · GPT-5.4 Mini XHigh Thinking · GPT-5.4 No Thinking · GPT-5.4 No Thinking Fast · GPT-5.4 XHigh Thinking · GPT-5.4 XHigh Thinking Fast · GPT-5.5 High Thinking · GPT-5.5 High Thinking Fast · GPT-5.5 Low Thinking · GPT-5.5 Low Thinking Fast · GPT-5.5 Medium Thinking · GPT-5.5 Medium Thinking Fast · GPT-5.5 No Thinking · GPT-5.5 No Thinking Fast · GPT-5.5 Review · GPT-5.5 XHigh Thinking · GPT-5.5 XHigh Thinking Fast · GPT-5.6 Luna High Thinking · GPT-5.6 Luna High Thinking Fast · GPT-5.6 Luna Low Thinking · GPT-5.6 Luna Low Thinking Fast · GPT-5.6 Luna Max Thinking · GPT-5.6 Luna Max Thinking Fast · GPT-5.6 Luna Medium Thinking · GPT-5.6 Luna Medium Thinking Fast · GPT-5.6 Luna No Thinking · GPT-5.6 Luna No Thinking Fast · GPT-5.6 Luna XHigh Thinking · GPT-5.6 Luna XHigh Thinking Fast · GPT-5.6 Sol High Thinking · GPT-5.6 Sol High Thinking Fast · GPT-5.6 Sol Low Thinking · GPT-5.6 Sol Low Thinking Fast · GPT-5.6 Sol Max Thinking · GPT-5.6 Sol Max Thinking Fast · GPT-5.6 Sol Medium Thinking · GPT-5.6 Sol Medium Thinking Fast · GPT-5.6 Sol No Thinking · GPT-5.6 Sol No Thinking Fast · GPT-5.6 Sol XHigh Thinking · GPT-5.6 Sol XHigh Thinking Fast · GPT-5.6 Terra High Thinking · GPT-5.6 Terra High Thinking Fast · GPT-5.6 Terra Low Thinking · GPT-5.6 Terra Low Thinking Fast · GPT-5.6 Terra Max Thinking · GPT-5.6 Terra Max Thinking Fast · GPT-5.6 Terra Medium Thinking · GPT-5.6 Terra Medium Thinking Fast · GPT-5.6 Terra No Thinking · GPT-5.6 Terra No Thinking Fast · GPT-5.6 Terra XHigh Thinking · GPT-5.6 Terra XHigh Thinking Fast · GPT-6 Astra High Thinking · GPT-6 Astra Low Thinking · GPT-6 Astra Max Thinking · GPT-6 Astra Medium Thinking · GPT-6 Astra XHigh Thinking · GPT-6 Astra High Thinking Fast · GPT-6 Astra Low Thinking Fast · GPT-6 Astra Medium Thinking Fast · GPT-6 Luna High Thinking · GPT-6 Luna High Thinking Fast · GPT-6 Luna Low Thinking · GPT-6 Luna Low Thinking Fast · GPT-6 Luna Max Thinking · GPT-6 Luna Max Thinking Fast · GPT-6 Luna Medium Thinking · GPT-6 Luna Medium Thinking Fast · GPT-6 Luna No Thinking · GPT-6 Luna No Thinking Fast · GPT-6 Luna XHigh Thinking · GPT-6 Luna XHigh Thinking Fast · GPT-6 Sol High Thinking · GPT-6 Sol High Thinking Fast · GPT-6 Sol Low Thinking · GPT-6 Sol Low Thinking Fast · GPT-6 Sol Max Thinking · GPT-6 Sol Max Thinking Fast · GPT-6 Sol Medium Thinking · GPT-6 Sol Medium Thinking Fast · GPT-6 Sol No Thinking · GPT-6 Sol No Thinking Fast · GPT-6 Sol XHigh Thinking · GPT-6 Sol XHigh Thinking Fast · GPT-6.1 Sol High Thinking · GPT-6.1 Sol High Thinking Fast · GPT-6.1 Sol Low Thinking · GPT-6.1 Sol Low Thinking Fast · GPT-6.1 Sol Max Thinking · GPT-6.1 Sol Max Thinking Fast · GPT-6.1 Sol Medium Thinking · GPT-6.1 Sol Medium Thinking Fast · GPT-6.1 Sol XHigh Thinking · GPT-6.1 Sol XHigh Thinking Fast · Grok 4.5 High · Grok 4.5 Low · Grok 4.5 Medium · Grok 4.6 High · Grok 4.6 Low · Grok 4.6 Medium · Grok 4.6 XHigh · Grok 4.7 High · Grok 4.7 Low · Grok 4.7 Medium · Grok 4.7 XHigh · Inkling High · Inkling Low · Inkling Max · Inkling Medium · Inkling None · Inkling X-High · Kimi K2.6 · Kimi K2.7 · Kimi K3 High · Kimi K3 Low · Kimi K3 Max · Nemotron 3 Ultra High · Nemotron 3 Ultra Medium · Nemotron 3 Ultra None · Opus 4.7 Review · SWE-1.6 · SWE-1.6 Fast · SWE-1.7 Max · SWE-1.7 Lightning Max · SWE-1.7 Lightning Medium · SWE-1.7 Medium · SWE-2 High · SWE-2 Max · SWE-2 Medium · SWE-check · xAI Grok-3 · xAI Grok-3 mini Thinking · DeepSeek V4 Flash High · DeepSeek V4 Flash Low · DeepSeek V4 Flash Max · DeepSeek V4 Pro Low · DeepSeek V4 Pro Max · DeepSeek V4.1 Flash High · DeepSeek V4.1 Flash Max · GLM-5.3 Flash High · GLM-5.3 Flash Low · GLM-5.3 Flash Max · GLM-5.3 High · GLM-5.3 Low · GLM-5.3 Max · GPT-6 Astra Max Thinking Fast · GPT-6 Astra XHigh Thinking Fast
A market label is not a guarantee of availability. Registration, payment and network conditions still apply.
Devin Desktop
Desktop · Devin Desktop
International
Mainland China support unverified
Formerly Windsurf, with an IDE, completion, ACP agent management and MCP extensions.
Coding
Agents
Source checked: 2026-10-11 Source verification does not certify the installer.
Full seats include Pro-equivalent daily/weekly quota and Desktop. Flex seats include no quota or Desktop; all usage draws from shared on-demand credits.
Devin AI 软件工程平台 · GPT-4o · GPT-4.1 · o3 · o3 High Reasoning · Claude Opus 4.5 · Claude Opus 4.5 Thinking · Gemini 2.5 Pro · Gemini 3 Flash High · Gemini 3 Flash Low · Gemini 3 Flash Medium · Gemini 3 Flash Minimal · GPT-5.2 High Thinking · GPT-5.2 High Thinking Fast · GPT-5.2 Low Thinking · GPT-5.2 Low Thinking Fast · GPT-5.2 Medium Thinking · GPT-5.2 Medium Thinking Fast · GPT-5.2 No Thinking · GPT-5.2 No Thinking Fast · GPT-5.2 XHigh Thinking · GPT-5.2 XHigh Thinking Fast · GPT-OSS 120B Medium Thinking · Claude Haiku 4.5 · GPT-5.1 No Thinking · GPT-5.1 Low Thinking · GPT-5.1 Medium Thinking · GPT-5.1 High Thinking · GPT-5.1 No Thinking Fast · GPT-5.1 Low Thinking Fast · GPT-5.1 Medium Thinking Fast · GPT-5.1 High Thinking Fast · Grok Code Fast 1 · GPT-5 Low Thinking · GPT-5 Medium Thinking · GPT-5 High Thinking · Adaptive · Fast Arena · Hybrid Arena · Frontier Arena · Claude Fable 5 High · Claude Fable 5 Low · Claude Fable 5 Max · Claude Fable 5 Medium · Claude Fable 5 XHigh · Claude Fable 5.1 High · Claude Fable 5.1 Low · Claude Fable 5.1 Max · Claude Fable 5.1 Medium · Claude Fable 5.1 XHigh · Claude Opus 4.6 · Claude Opus 4.6 1M · Claude Opus 4.6 Thinking · Claude Opus 4.6 Thinking 1M · Claude Opus 4.7 High · Claude Opus 4.7 Low · Claude Opus 4.7 Max · Claude Opus 4.7 Medium · Claude Opus 4.7 XHigh · Claude Opus 4.8 High · Claude Opus 4.8 High Fast · Claude Opus 4.8 Low · Claude Opus 4.8 Low Fast · Claude Opus 4.8 Max · Claude Opus 4.8 Max Fast · Claude Opus 4.8 Medium · Claude Opus 4.8 Medium Fast · Claude Opus 4.8 XHigh · Claude Opus 4.8 XHigh Fast · Claude Opus 5 High · Claude Opus 5 High Fast · Claude Opus 5 Low · Claude Opus 5 Low Fast · Claude Opus 5 Max · Claude Opus 5 Max Fast · Claude Opus 5 Medium · Claude Opus 5 Medium Fast · Claude Opus 5 XHigh · Claude Opus 5 XHigh Fast · Claude Opus 5.5 High · Claude Opus 5.5 High Fast · Claude Opus 5.5 Low · Claude Opus 5.5 Low Fast · Claude Opus 5.5 Max · Claude Opus 5.5 Max Fast · Claude Opus 5.5 Medium · Claude Opus 5.5 Medium Fast · Claude Opus 5.5 XHigh · Claude Opus 5.5 XHigh Fast · Claude Sonnet 4.6 · Claude Sonnet 4.6 1M · Claude Sonnet 4.6 Thinking · Claude Sonnet 4.6 Thinking 1M · Claude Sonnet 5 High · Claude Sonnet 5 Low · Claude Sonnet 5 Max · Claude Sonnet 5 Medium · Claude Sonnet 5 XHigh · Claude Sonnet 5.5 High · Claude Sonnet 5.5 Low · Claude Sonnet 5.5 Max · Claude Sonnet 5.5 Medium · Claude Sonnet 5.5 XHigh · DeepSeek V4 Pro High · Gemini 3.1 Pro High Thinking · Gemini 3.1 Pro Low Thinking · Gemini 3.5 Flash High · Gemini 3.5 Flash Low · Gemini 3.5 Flash Medium · Gemini 3.5 Flash Minimal · Gemini 3.6 Flash High · Gemini 3.6 Flash Low · Gemini 3.6 Flash Medium · Gemini 3.6 Flash Minimal · Gemini 3.7 Flash High · Gemini 3.7 Flash Low · Gemini 3.7 Flash Medium · Gemini 3.8 Flash High · Gemini 3.8 Flash Low · Gemini 3.8 Flash Medium · GLM-5.2 High · GLM-5.2 High 1M · GLM-5.2 Max · GLM-5.2 Max 1M · GLM-5.2 No Thinking · GLM-5.2 No Thinking 1M · GPT-5.3-Codex High · GPT-5.3-Codex High Fast · GPT-5.3-Codex Low · GPT-5.3-Codex Low Fast · GPT-5.3-Codex Medium · GPT-5.3-Codex Medium Fast · GPT-5.3-Codex Spark Medium · GPT-5.3-Codex X-High · GPT-5.3-Codex XHigh Fast · GPT-5.4 High Thinking · GPT-5.4 High Thinking Fast · GPT-5.4 Low Thinking · GPT-5.4 Low Thinking Fast · GPT-5.4 Medium Thinking · GPT-5.4 Medium Thinking Fast · GPT-5.4 Mini High Thinking · GPT-5.4 Mini Low Thinking · GPT-5.4 Mini Medium Thinking · GPT-5.4 Mini XHigh Thinking · GPT-5.4 No Thinking · GPT-5.4 No Thinking Fast · GPT-5.4 XHigh Thinking · GPT-5.4 XHigh Thinking Fast · GPT-5.5 High Thinking · GPT-5.5 High Thinking Fast · GPT-5.5 Low Thinking · GPT-5.5 Low Thinking Fast · GPT-5.5 Medium Thinking · GPT-5.5 Medium Thinking Fast · GPT-5.5 No Thinking · GPT-5.5 No Thinking Fast · GPT-5.5 Review · GPT-5.5 XHigh Thinking · GPT-5.5 XHigh Thinking Fast · GPT-5.6 Luna High Thinking · GPT-5.6 Luna High Thinking Fast · GPT-5.6 Luna Low Thinking · GPT-5.6 Luna Low Thinking Fast · GPT-5.6 Luna Max Thinking · GPT-5.6 Luna Max Thinking Fast · GPT-5.6 Luna Medium Thinking · GPT-5.6 Luna Medium Thinking Fast · GPT-5.6 Luna No Thinking · GPT-5.6 Luna No Thinking Fast · GPT-5.6 Luna XHigh Thinking · GPT-5.6 Luna XHigh Thinking Fast · GPT-5.6 Sol High Thinking · GPT-5.6 Sol High Thinking Fast · GPT-5.6 Sol Low Thinking · GPT-5.6 Sol Low Thinking Fast · GPT-5.6 Sol Max Thinking · GPT-5.6 Sol Max Thinking Fast · GPT-5.6 Sol Medium Thinking · GPT-5.6 Sol Medium Thinking Fast · GPT-5.6 Sol No Thinking · GPT-5.6 Sol No Thinking Fast · GPT-5.6 Sol XHigh Thinking · GPT-5.6 Sol XHigh Thinking Fast · GPT-5.6 Terra High Thinking · GPT-5.6 Terra High Thinking Fast · GPT-5.6 Terra Low Thinking · GPT-5.6 Terra Low Thinking Fast · GPT-5.6 Terra Max Thinking · GPT-5.6 Terra Max Thinking Fast · GPT-5.6 Terra Medium Thinking · GPT-5.6 Terra Medium Thinking Fast · GPT-5.6 Terra No Thinking · GPT-5.6 Terra No Thinking Fast · GPT-5.6 Terra XHigh Thinking · GPT-5.6 Terra XHigh Thinking Fast · GPT-6 Astra High Thinking · GPT-6 Astra Low Thinking · GPT-6 Astra Max Thinking · GPT-6 Astra Medium Thinking · GPT-6 Astra XHigh Thinking · GPT-6 Astra High Thinking Fast · GPT-6 Astra Low Thinking Fast · GPT-6 Astra Medium Thinking Fast · GPT-6 Luna High Thinking · GPT-6 Luna High Thinking Fast · GPT-6 Luna Low Thinking · GPT-6 Luna Low Thinking Fast · GPT-6 Luna Max Thinking · GPT-6 Luna Max Thinking Fast · GPT-6 Luna Medium Thinking · GPT-6 Luna Medium Thinking Fast · GPT-6 Luna No Thinking · GPT-6 Luna No Thinking Fast · GPT-6 Luna XHigh Thinking · GPT-6 Luna XHigh Thinking Fast · GPT-6 Sol High Thinking · GPT-6 Sol High Thinking Fast · GPT-6 Sol Low Thinking · GPT-6 Sol Low Thinking Fast · GPT-6 Sol Max Thinking · GPT-6 Sol Max Thinking Fast · GPT-6 Sol Medium Thinking · GPT-6 Sol Medium Thinking Fast · GPT-6 Sol No Thinking · GPT-6 Sol No Thinking Fast · GPT-6 Sol XHigh Thinking · GPT-6 Sol XHigh Thinking Fast · GPT-6.1 Sol High Thinking · GPT-6.1 Sol High Thinking Fast · GPT-6.1 Sol Low Thinking · GPT-6.1 Sol Low Thinking Fast · GPT-6.1 Sol Max Thinking · GPT-6.1 Sol Max Thinking Fast · GPT-6.1 Sol Medium Thinking · GPT-6.1 Sol Medium Thinking Fast · GPT-6.1 Sol XHigh Thinking · GPT-6.1 Sol XHigh Thinking Fast · Grok 4.5 High · Grok 4.5 Low · Grok 4.5 Medium · Grok 4.6 High · Grok 4.6 Low · Grok 4.6 Medium · Grok 4.6 XHigh · Grok 4.7 High · Grok 4.7 Low · Grok 4.7 Medium · Grok 4.7 XHigh · Inkling High · Inkling Low · Inkling Max · Inkling Medium · Inkling None · Inkling X-High · Kimi K2.6 · Kimi K2.7 · Kimi K3 High · Kimi K3 Low · Kimi K3 Max · Nemotron 3 Ultra High · Nemotron 3 Ultra Medium · Nemotron 3 Ultra None · Opus 4.7 Review · SWE-1.6 · SWE-1.6 Fast · SWE-1.7 Max · SWE-1.7 Lightning Max · SWE-1.7 Lightning Medium · SWE-1.7 Medium · SWE-2 High · SWE-2 Max · SWE-2 Medium · SWE-check · xAI Grok-3 · xAI Grok-3 mini Thinking · DeepSeek V4 Flash High · DeepSeek V4 Flash Low · DeepSeek V4 Flash Max · DeepSeek V4 Pro Low · DeepSeek V4 Pro Max · DeepSeek V4.1 Flash High · DeepSeek V4.1 Flash Max · GLM-5.3 Flash High · GLM-5.3 Flash Low · GLM-5.3 Flash Max · GLM-5.3 High · GLM-5.3 Low · GLM-5.3 Max · GPT-6 Astra Max Thinking Fast · GPT-6 Astra XHigh Thinking Fast
A market label is not a guarantee of availability. Registration, payment and network conditions still apply.
Devin CLI
CLI · Devin CLI
International
Mainland China support unverified
A local terminal coding agent with model switching, Fusion, MCP, Skills and Cloud Handoff.
Coding
Agents
Source checked: 2026-10-11 Source verification does not certify the installer.
Full seats include Pro-equivalent daily/weekly quota and Desktop. Flex seats include no quota or Desktop; all usage draws from shared on-demand credits.
Devin AI 软件工程平台 · GPT-4o · GPT-4.1 · o3 · o3 High Reasoning · Claude Opus 4.5 · Claude Opus 4.5 Thinking · Gemini 2.5 Pro · Gemini 3 Flash High · Gemini 3 Flash Low · Gemini 3 Flash Medium · Gemini 3 Flash Minimal · GPT-5.2 High Thinking · GPT-5.2 High Thinking Fast · GPT-5.2 Low Thinking · GPT-5.2 Low Thinking Fast · GPT-5.2 Medium Thinking · GPT-5.2 Medium Thinking Fast · GPT-5.2 No Thinking · GPT-5.2 No Thinking Fast · GPT-5.2 XHigh Thinking · GPT-5.2 XHigh Thinking Fast · GPT-OSS 120B Medium Thinking · Claude Haiku 4.5 · GPT-5.1 No Thinking · GPT-5.1 Low Thinking · GPT-5.1 Medium Thinking · GPT-5.1 High Thinking · GPT-5.1 No Thinking Fast · GPT-5.1 Low Thinking Fast · GPT-5.1 Medium Thinking Fast · GPT-5.1 High Thinking Fast · Grok Code Fast 1 · GPT-5 Low Thinking · GPT-5 Medium Thinking · GPT-5 High Thinking · Adaptive · Fast Arena · Hybrid Arena · Frontier Arena · Claude Fable 5 High · Claude Fable 5 Low · Claude Fable 5 Max · Claude Fable 5 Medium · Claude Fable 5 XHigh · Claude Fable 5.1 High · Claude Fable 5.1 Low · Claude Fable 5.1 Max · Claude Fable 5.1 Medium · Claude Fable 5.1 XHigh · Claude Opus 4.6 · Claude Opus 4.6 1M · Claude Opus 4.6 Thinking · Claude Opus 4.6 Thinking 1M · Claude Opus 4.7 High · Claude Opus 4.7 Low · Claude Opus 4.7 Max · Claude Opus 4.7 Medium · Claude Opus 4.7 XHigh · Claude Opus 4.8 High · Claude Opus 4.8 High Fast · Claude Opus 4.8 Low · Claude Opus 4.8 Low Fast · Claude Opus 4.8 Max · Claude Opus 4.8 Max Fast · Claude Opus 4.8 Medium · Claude Opus 4.8 Medium Fast · Claude Opus 4.8 XHigh · Claude Opus 4.8 XHigh Fast · Claude Opus 5 High · Claude Opus 5 High Fast · Claude Opus 5 Low · Claude Opus 5 Low Fast · Claude Opus 5 Max · Claude Opus 5 Max Fast · Claude Opus 5 Medium · Claude Opus 5 Medium Fast · Claude Opus 5 XHigh · Claude Opus 5 XHigh Fast · Claude Opus 5.5 High · Claude Opus 5.5 High Fast · Claude Opus 5.5 Low · Claude Opus 5.5 Low Fast · Claude Opus 5.5 Max · Claude Opus 5.5 Max Fast · Claude Opus 5.5 Medium · Claude Opus 5.5 Medium Fast · Claude Opus 5.5 XHigh · Claude Opus 5.5 XHigh Fast · Claude Sonnet 4.6 · Claude Sonnet 4.6 1M · Claude Sonnet 4.6 Thinking · Claude Sonnet 4.6 Thinking 1M · Claude Sonnet 5 High · Claude Sonnet 5 Low · Claude Sonnet 5 Max · Claude Sonnet 5 Medium · Claude Sonnet 5 XHigh · Claude Sonnet 5.5 High · Claude Sonnet 5.5 Low · Claude Sonnet 5.5 Max · Claude Sonnet 5.5 Medium · Claude Sonnet 5.5 XHigh · DeepSeek V4 Pro High · Gemini 3.1 Pro High Thinking · Gemini 3.1 Pro Low Thinking · Gemini 3.5 Flash High · Gemini 3.5 Flash Low · Gemini 3.5 Flash Medium · Gemini 3.5 Flash Minimal · Gemini 3.6 Flash High · Gemini 3.6 Flash Low · Gemini 3.6 Flash Medium · Gemini 3.6 Flash Minimal · Gemini 3.7 Flash High · Gemini 3.7 Flash Low · Gemini 3.7 Flash Medium · Gemini 3.8 Flash High · Gemini 3.8 Flash Low · Gemini 3.8 Flash Medium · GLM-5.2 High · GLM-5.2 High 1M · GLM-5.2 Max · GLM-5.2 Max 1M · GLM-5.2 No Thinking · GLM-5.2 No Thinking 1M · GPT-5.3-Codex High · GPT-5.3-Codex High Fast · GPT-5.3-Codex Low · GPT-5.3-Codex Low Fast · GPT-5.3-Codex Medium · GPT-5.3-Codex Medium Fast · GPT-5.3-Codex Spark Medium · GPT-5.3-Codex X-High · GPT-5.3-Codex XHigh Fast · GPT-5.4 High Thinking · GPT-5.4 High Thinking Fast · GPT-5.4 Low Thinking · GPT-5.4 Low Thinking Fast · GPT-5.4 Medium Thinking · GPT-5.4 Medium Thinking Fast · GPT-5.4 Mini High Thinking · GPT-5.4 Mini Low Thinking · GPT-5.4 Mini Medium Thinking · GPT-5.4 Mini XHigh Thinking · GPT-5.4 No Thinking · GPT-5.4 No Thinking Fast · GPT-5.4 XHigh Thinking · GPT-5.4 XHigh Thinking Fast · GPT-5.5 High Thinking · GPT-5.5 High Thinking Fast · GPT-5.5 Low Thinking · GPT-5.5 Low Thinking Fast · GPT-5.5 Medium Thinking · GPT-5.5 Medium Thinking Fast · GPT-5.5 No Thinking · GPT-5.5 No Thinking Fast · GPT-5.5 Review · GPT-5.5 XHigh Thinking · GPT-5.5 XHigh Thinking Fast · GPT-5.6 Luna High Thinking · GPT-5.6 Luna High Thinking Fast · GPT-5.6 Luna Low Thinking · GPT-5.6 Luna Low Thinking Fast · GPT-5.6 Luna Max Thinking · GPT-5.6 Luna Max Thinking Fast · GPT-5.6 Luna Medium Thinking · GPT-5.6 Luna Medium Thinking Fast · GPT-5.6 Luna No Thinking · GPT-5.6 Luna No Thinking Fast · GPT-5.6 Luna XHigh Thinking · GPT-5.6 Luna XHigh Thinking Fast · GPT-5.6 Sol High Thinking · GPT-5.6 Sol High Thinking Fast · GPT-5.6 Sol Low Thinking · GPT-5.6 Sol Low Thinking Fast · GPT-5.6 Sol Max Thinking · GPT-5.6 Sol Max Thinking Fast · GPT-5.6 Sol Medium Thinking · GPT-5.6 Sol Medium Thinking Fast · GPT-5.6 Sol No Thinking · GPT-5.6 Sol No Thinking Fast · GPT-5.6 Sol XHigh Thinking · GPT-5.6 Sol XHigh Thinking Fast · GPT-5.6 Terra High Thinking · GPT-5.6 Terra High Thinking Fast · GPT-5.6 Terra Low Thinking · GPT-5.6 Terra Low Thinking Fast · GPT-5.6 Terra Max Thinking · GPT-5.6 Terra Max Thinking Fast · GPT-5.6 Terra Medium Thinking · GPT-5.6 Terra Medium Thinking Fast · GPT-5.6 Terra No Thinking · GPT-5.6 Terra No Thinking Fast · GPT-5.6 Terra XHigh Thinking · GPT-5.6 Terra XHigh Thinking Fast · GPT-6 Astra High Thinking · GPT-6 Astra Low Thinking · GPT-6 Astra Max Thinking · GPT-6 Astra Medium Thinking · GPT-6 Astra XHigh Thinking · GPT-6 Astra High Thinking Fast · GPT-6 Astra Low Thinking Fast · GPT-6 Astra Medium Thinking Fast · GPT-6 Luna High Thinking · GPT-6 Luna High Thinking Fast · GPT-6 Luna Low Thinking · GPT-6 Luna Low Thinking Fast · GPT-6 Luna Max Thinking · GPT-6 Luna Max Thinking Fast · GPT-6 Luna Medium Thinking · GPT-6 Luna Medium Thinking Fast · GPT-6 Luna No Thinking · GPT-6 Luna No Thinking Fast · GPT-6 Luna XHigh Thinking · GPT-6 Luna XHigh Thinking Fast · GPT-6 Sol High Thinking · GPT-6 Sol High Thinking Fast · GPT-6 Sol Low Thinking · GPT-6 Sol Low Thinking Fast · GPT-6 Sol Max Thinking · GPT-6 Sol Max Thinking Fast · GPT-6 Sol Medium Thinking · GPT-6 Sol Medium Thinking Fast · GPT-6 Sol No Thinking · GPT-6 Sol No Thinking Fast · GPT-6 Sol XHigh Thinking · GPT-6 Sol XHigh Thinking Fast · GPT-6.1 Sol High Thinking · GPT-6.1 Sol High Thinking Fast · GPT-6.1 Sol Low Thinking · GPT-6.1 Sol Low Thinking Fast · GPT-6.1 Sol Max Thinking · GPT-6.1 Sol Max Thinking Fast · GPT-6.1 Sol Medium Thinking · GPT-6.1 Sol Medium Thinking Fast · GPT-6.1 Sol XHigh Thinking · GPT-6.1 Sol XHigh Thinking Fast · Grok 4.5 High · Grok 4.5 Low · Grok 4.5 Medium · Grok 4.6 High · Grok 4.6 Low · Grok 4.6 Medium · Grok 4.6 XHigh · Grok 4.7 High · Grok 4.7 Low · Grok 4.7 Medium · Grok 4.7 XHigh · Inkling High · Inkling Low · Inkling Max · Inkling Medium · Inkling None · Inkling X-High · Kimi K2.6 · Kimi K2.7 · Kimi K3 High · Kimi K3 Low · Kimi K3 Max · Nemotron 3 Ultra High · Nemotron 3 Ultra Medium · Nemotron 3 Ultra None · Opus 4.7 Review · SWE-1.6 · SWE-1.6 Fast · SWE-1.7 Max · SWE-1.7 Lightning Max · SWE-1.7 Lightning Medium · SWE-1.7 Medium · SWE-2 High · SWE-2 Max · SWE-2 Medium · SWE-check · xAI Grok-3 · xAI Grok-3 mini Thinking · DeepSeek V4 Flash High · DeepSeek V4 Flash Low · DeepSeek V4 Flash Max · DeepSeek V4 Pro Low · DeepSeek V4 Pro Max · DeepSeek V4.1 Flash High · DeepSeek V4.1 Flash Max · GLM-5.3 Flash High · GLM-5.3 Flash Low · GLM-5.3 Flash Max · GLM-5.3 High · GLM-5.3 Low · GLM-5.3 Max · GPT-6 Astra Max Thinking Fast · GPT-6 Astra XHigh Thinking Fast
A market label is not a guarantee of availability. Registration, payment and network conditions still apply.
Kiro IDE
Desktop · Kiro IDE
International
Conditionally supported in mainland China
An AI coding platform for spec-driven development, MCP, hooks and agent workflows.
Coding
Agents
Source checked: 2026-10-12 Source verification does not certify the installer.
Actual consumption depends on the task and model; simple work may use less than one credit and complex work several. Credits are not request counts or tokens and are not interchangeable with other vendors' or CreditsPlan credits.
Unused monthly subscription credits do not roll over.
Subscription credits reset at the start of each calendar month.
Add-on credits (a separate purchase) — USD/Credit: 0.04; minimum purchase USD: 5; minimum Credits: 125; maximum pack USD: 100; maximum active packs: 5; validity in months: 12
Unused add-on credits roll over within their validity period.
Use monthly subscription credits first, then add-on credits.
This subscription does not establish a general OpenAI-compatible model API.
Subscriptions cannot be assumed usable with arbitrary third-party tools; third-party harnesses such as OpenClaw are not permitted.
Official CLI API keys support documented headless development automation. A paid plan is required; the Free plan does not include API-key authentication. Enterprise accounts require administrator enablement. Official authentication conditions
Models and Credits multipliers · 22
Multipliers measure relative platform usage consumption, not API token prices. Billing countries, model service regions and per-plan model access are separate conditions.
Country, account, provider and model selection restrictions apply; billing-country eligibility does not guarantee model access.
Up to $20 toward the first paid subscription, prorated for a mid-month upgrade. Requires social login or AWS Builder ID and a valid credit card; excludes IAM Identity Center. Not a grant of Kiro Credits.
Applicable plans: Kiro Pro · Kiro Pro+ · Kiro Pro Max · Kiro Power
Conditions: First paid upgrade · Social login or AWS Builder ID · Valid credit card
No end date published; confirm eligibility on the official page.
Kiro startup credit application program
Approved early-stage through Series A startups can receive up to one year of Pro+ credits for 2/10/30 users. Requires IAM Identity Center; regional exclusions including China/GCR and AWS Activate restrictions apply. Standard billing follows credit exhaustion.
Applicable plans: Kiro Pro+
Conditions: Early-stage through Series A startup · Matching business domain and AWS account · Approved application · AWS IAM Identity Center
Excluded sign-in methods: Social login · AWS Builder ID
Excluded regions: Argentina · Belarus · Brazil · China/GCR · Cuba · France · Germany · Iran · Italy · Mexico · North Korea · Poland · Russia · Spain · Syria · Crimea · DNR · LNR · United Arab Emirates
Actual consumption depends on the task and model; simple work may use less than one credit and complex work several. Credits are not request counts or tokens and are not interchangeable with other vendors' or CreditsPlan credits.
Unused monthly subscription credits do not roll over.
Subscription credits reset at the start of each calendar month.
Add-on credits (a separate purchase) — USD/Credit: 0.04; minimum purchase USD: 5; minimum Credits: 125; maximum pack USD: 100; maximum active packs: 5; validity in months: 12
Unused add-on credits roll over within their validity period.
Use monthly subscription credits first, then add-on credits.
This subscription does not establish a general OpenAI-compatible model API.
Subscriptions cannot be assumed usable with arbitrary third-party tools; third-party harnesses such as OpenClaw are not permitted.
Official CLI API keys support documented headless development automation. A paid plan is required; the Free plan does not include API-key authentication. Enterprise accounts require administrator enablement. Official authentication conditions
Models and Credits multipliers · 22
Multipliers measure relative platform usage consumption, not API token prices. Billing countries, model service regions and per-plan model access are separate conditions.
Country, account, provider and model selection restrictions apply; billing-country eligibility does not guarantee model access.
Up to $20 toward the first paid subscription, prorated for a mid-month upgrade. Requires social login or AWS Builder ID and a valid credit card; excludes IAM Identity Center. Not a grant of Kiro Credits.
Applicable plans: Kiro Pro · Kiro Pro+ · Kiro Pro Max · Kiro Power
Conditions: First paid upgrade · Social login or AWS Builder ID · Valid credit card
No end date published; confirm eligibility on the official page.
Kiro startup credit application program
Approved early-stage through Series A startups can receive up to one year of Pro+ credits for 2/10/30 users. Requires IAM Identity Center; regional exclusions including China/GCR and AWS Activate restrictions apply. Standard billing follows credit exhaustion.
Applicable plans: Kiro Pro+
Conditions: Early-stage through Series A startup · Matching business domain and AWS account · Approved application · AWS IAM Identity Center
Excluded sign-in methods: Social login · AWS Builder ID
Excluded regions: Argentina · Belarus · Brazil · China/GCR · Cuba · France · Germany · Iran · Italy · Mexico · North Korea · Poland · Russia · Spain · Syria · Crimea · DNR · LNR · United Arab Emirates
Actual consumption depends on the task and model; simple work may use less than one credit and complex work several. Credits are not request counts or tokens and are not interchangeable with other vendors' or CreditsPlan credits.
Unused monthly subscription credits do not roll over.
Subscription credits reset at the start of each calendar month.
Add-on credits (a separate purchase) — USD/Credit: 0.04; minimum purchase USD: 5; minimum Credits: 125; maximum pack USD: 100; maximum active packs: 5; validity in months: 12
Unused add-on credits roll over within their validity period.
Use monthly subscription credits first, then add-on credits.
This subscription does not establish a general OpenAI-compatible model API.
Subscriptions cannot be assumed usable with arbitrary third-party tools; third-party harnesses such as OpenClaw are not permitted.
Official CLI API keys support documented headless development automation. A paid plan is required; the Free plan does not include API-key authentication. Enterprise accounts require administrator enablement. Official authentication conditions
Models and Credits multipliers · 22
Multipliers measure relative platform usage consumption, not API token prices. Billing countries, model service regions and per-plan model access are separate conditions.
Country, account, provider and model selection restrictions apply; billing-country eligibility does not guarantee model access.
Up to $20 toward the first paid subscription, prorated for a mid-month upgrade. Requires social login or AWS Builder ID and a valid credit card; excludes IAM Identity Center. Not a grant of Kiro Credits.
Applicable plans: Kiro Pro · Kiro Pro+ · Kiro Pro Max · Kiro Power
Conditions: First paid upgrade · Social login or AWS Builder ID · Valid credit card
No end date published; confirm eligibility on the official page.
Kiro startup credit application program
Approved early-stage through Series A startups can receive up to one year of Pro+ credits for 2/10/30 users. Requires IAM Identity Center; regional exclusions including China/GCR and AWS Activate restrictions apply. Standard billing follows credit exhaustion.
Applicable plans: Kiro Pro+
Conditions: Early-stage through Series A startup · Matching business domain and AWS account · Approved application · AWS IAM Identity Center
Excluded sign-in methods: Social login · AWS Builder ID
Excluded regions: Argentina · Belarus · Brazil · China/GCR · Cuba · France · Germany · Iran · Italy · Mexico · North Korea · Poland · Russia · Spain · Syria · Crimea · DNR · LNR · United Arab Emirates
No end date published; confirm eligibility on the official page.
A market label is not a guarantee of availability. Registration, payment and network conditions still apply.
MiniMax Design
Desktop
China
International
Mainland China support unverified
MiniMax’s desktop multimodal creative agent combines task planning, a canvas, Skills and local assets to create images, video, music, voiceovers and brand marketing content.
AI design
Multimodal creation
Image creation and editing
Video creation
Audio and voiceover
Marketing
E-commerce assets
Agents
Source checked: 2026-10-12 Source verification does not certify the installer.
Agent task planning and multimodal creative workflows
Image and video editing with a canvas
Skills plugins, local assets and professional tool connections
Suitable tasks: E-commerce product visuals and batch marketing images; E-commerce advertising videos and short-form video creation; Brand promotional content and advertising assets; Creative projects combining images, video and audio; Short drama production
Product pricing · CNY and USD plans are listed separately by market. Subscription Credits are granted monthly with 31-day validity; purchased Credits last one year. Media tasks are pre-deducted; Agent/LLM tasks are charged after actual Token usage. Free allowance amounts, specific task consumption and commercial-use rights remain unverified. Design Credits are separate from CreditsPlan chat Credits and MiniMax Code/API quotas.
Starter · China · CNY 65/month · CNY 624/year
10,000 Credits / month
Each subscription Credits grant is valid for 31 days; rollover is not explicitly documented.; Specific task Credits consumption, exact concurrency, commercial-use rights and regional payment conditions are not yet verified.
Each subscription Credits grant is valid for 31 days; rollover is not explicitly documented.; Specific task Credits consumption, exact concurrency, commercial-use rights and regional payment conditions are not yet verified.
Each subscription Credits grant is valid for 31 days; rollover is not explicitly documented.; Specific task Credits consumption, exact concurrency, commercial-use rights and regional payment conditions are not yet verified.
Credits closest to expiry are deducted first; other stacking and rollover details are not yet verified.; Specific task Credits consumption, commercial-use rights and regional payment conditions are not yet verified.
Credits closest to expiry are deducted first; other stacking and rollover details are not yet verified.; Specific task Credits consumption, commercial-use rights and regional payment conditions are not yet verified.
Credits closest to expiry are deducted first; other stacking and rollover details are not yet verified.; Specific task Credits consumption, commercial-use rights and regional payment conditions are not yet verified.
Credits closest to expiry are deducted first; other stacking and rollover details are not yet verified.; Specific task Credits consumption, commercial-use rights and regional payment conditions are not yet verified.
Starter · International · USD 10.5/month · USD 100.8/year
10,000 Credits / month
Each subscription Credits grant is valid for 31 days; rollover is not explicitly documented.; Specific task Credits consumption, exact concurrency, commercial-use rights and regional payment conditions are not yet verified.
Plus · International · USD 76/month · USD 729.6/year
87,000 Credits / month
Each subscription Credits grant is valid for 31 days; rollover is not explicitly documented.; Specific task Credits consumption, exact concurrency, commercial-use rights and regional payment conditions are not yet verified.
Pro · International · USD 215/month · USD 2064/year
270,000 Credits / month
Each subscription Credits grant is valid for 31 days; rollover is not explicitly documented.; Specific task Credits consumption, exact concurrency, commercial-use rights and regional payment conditions are not yet verified.
Starter · International · USD 10 (one-time Credits pack)
9,250 Credits per purchase
Credits closest to expiry are deducted first; other stacking and rollover details are not yet verified.; Specific task Credits consumption, commercial-use rights and regional payment conditions are not yet verified.
Growth · International · USD 50 (one-time Credits pack)
46,250 Credits per purchase
Credits closest to expiry are deducted first; other stacking and rollover details are not yet verified.; Specific task Credits consumption, commercial-use rights and regional payment conditions are not yet verified.
High Volume · International · USD 200 (one-time Credits pack)
194,250 Credits per purchase
Credits closest to expiry are deducted first; other stacking and rollover details are not yet verified.; Specific task Credits consumption, commercial-use rights and regional payment conditions are not yet verified.
Pro · International · USD 500 (one-time Credits pack)
508,750 Credits per purchase
Credits closest to expiry are deducted first; other stacking and rollover details are not yet verified.; Specific task Credits consumption, commercial-use rights and regional payment conditions are not yet verified.
The official homepage describes multimodal creation, an Agent, a canvas, Skills and local assets. · Verified · 2026-10-12 · Official evidence
The international homepage presents image, video, audio and creative workflows. · Verified · 2026-10-12 · Official evidence
The official download entry states Windows 10+ x64 and macOS 13+; only stable official pages are retained. · Verified · 2026-10-12 · Official evidence
Design subscription and Credits-pack prices are verified for this market; checkout controls final price and applicable tax or fees. · Verified · 2026-10-12 · Official evidence
Verified subscription tiers for this product. · Verified · 2026-10-12 · Official evidence
Verified standalone Credits purchases for this product. · Verified · 2026-10-12 · Official evidence
Subscription Credits are monthly with 31-day validity; purchased Credits last one year; the earliest expiry is deducted first. Image, video and audio tasks are pre-deducted; Agent/LLM tasks are charged after actual Tokens usage. Design Credits do not transfer to other MiniMax products. · Verified · 2026-10-12 · Official evidence
A registered account is required. Subscriptions are for individuals and belong to the personal default space. Only the team Owner may purchase team Credits, which bind to the correct current team; verify the team or personal account before payment. · Verified · 2026-10-12 · Official evidence
Design subscription and Credits-pack prices are verified for this market; checkout controls final price and applicable tax or fees. · Verified · 2026-10-12 · Official evidence
Verified subscription tiers for this product. · Verified · 2026-10-12 · Official evidence
Verified standalone Credits purchases for this product. · Verified · 2026-10-12 · Official evidence
Subscription Credits are monthly with 31-day validity; purchased Credits last one year; the earliest expiry is deducted first. Image, video and audio tasks are pre-deducted; Agent/LLM tasks are charged after actual Tokens usage. Design Credits do not transfer to other MiniMax products. · Verified · 2026-10-12 · Official evidence
A registered account is required. Subscriptions are for individuals and belong to the personal default space. Only the team Owner may purchase team Credits, which bind to the correct current team; verify the team or personal account before payment. · Verified · 2026-10-12 · Official evidence
The official tools page offers a free start and mentions free Credits; amount, eligibility and cadence are undisclosed, so no free plan is created. · Verified · 2026-10-12 · Official evidence
A market label is not a guarantee of availability. Registration, payment and network conditions still apply.
Poe
Web
International
Mainland China support unverified
Quora multi-model AI platform for chat, media and custom bots.
Chat
Productivity
Source checked: 2026-10-12 Source verification does not certify the installer.
Poe Compute Points are an internal platform billing allowance, not equivalent cash or a fixed number of tokens. Usage is charged at the model's actual rate; a fixed message count or usable duration cannot be inferred.
Chat and the external-application model API share the account's points.
Paid models through the API require an active subscription or purchased add-on points; bot creation APIs and creator monetization are separate mechanisms.
Additional points can be purchased separately, subject to account pricing and eligibility.
Purchased points are valid for 12 months; their purchase price is not a redeemable cash exchange rate.
Platform bot families and providers · 5
Model availability, versions, context and point costs vary by bot and account. Free-plan access to these families remains unverified.
OpenAI GPT bots (OpenAI) · Anthropic Claude bots (Anthropic) · Google Gemini bots (Google) · FLUX and Ideogram bots (Black Forest Labs / Ideogram) · Runway, Luma, Hailuo and Pika bots (Runway / Luma / MiniMax / Pika)
A market label is not a guarantee of availability. Registration, payment and network conditions still apply.
Poe API
API
International
Mainland China support unverified
OpenAI-compatible API with Poe points; paid models require an active subscription or add-on points.
Model APIs
Chat
Source checked: 2026-10-12 Source verification does not certify the installer.
Poe Compute Points are an internal platform billing allowance, not equivalent cash or a fixed number of tokens. Usage is charged at the model's actual rate; a fixed message count or usable duration cannot be inferred.
Chat and the external-application model API share the account's points.
Paid models through the API require an active subscription or purchased add-on points; bot creation APIs and creator monetization are separate mechanisms.
Additional points can be purchased separately, subject to account pricing and eligibility.
Purchased points are valid for 12 months; their purchase price is not a redeemable cash exchange rate.
Platform bot families and providers · 5
Model availability, versions, context and point costs vary by bot and account. Free-plan access to these families remains unverified.
OpenAI GPT bots (OpenAI) · Anthropic Claude bots (Anthropic) · Google Gemini bots (Google) · FLUX and Ideogram bots (Black Forest Labs / Ideogram) · Runway, Luma, Hailuo and Pika bots (Runway / Luma / MiniMax / Pika)
A market label is not a guarantee of availability. Registration, payment and network conditions still apply.
Poe Desktop
Desktop
International
Mainland China support unverified
Official Poe Windows desktop acquisition page.
Chat
Productivity
Source checked: 2026-10-12 Source verification does not certify the installer.
Poe Compute Points are an internal platform billing allowance, not equivalent cash or a fixed number of tokens. Usage is charged at the model's actual rate; a fixed message count or usable duration cannot be inferred.
Chat and the external-application model API share the account's points.
Paid models through the API require an active subscription or purchased add-on points; bot creation APIs and creator monetization are separate mechanisms.
Additional points can be purchased separately, subject to account pricing and eligibility.
Purchased points are valid for 12 months; their purchase price is not a redeemable cash exchange rate.
Platform bot families and providers · 5
Model availability, versions, context and point costs vary by bot and account. Free-plan access to these families remains unverified.
OpenAI GPT bots (OpenAI) · Anthropic Claude bots (Anthropic) · Google Gemini bots (Google) · FLUX and Ideogram bots (Black Forest Labs / Ideogram) · Runway, Luma, Hailuo and Pika bots (Runway / Luma / MiniMax / Pika)
A market label is not a guarantee of availability. Registration, payment and network conditions still apply.
Chatbox AI
Desktop
China
International
Conditionally supported in mainland China
Chatbox AI is a multi-model AI chat client for desktop and mobile devices. It supports Chatbox-hosted subscriptions, your own provider API keys and local models; costs and available capabilities depend on the selected service mode and plan.
Chat
Productivity
Local models
Source checked: 2026-10-12 Source verification does not certify the installer.
Switch between AI models in one client; Use an existing API key and avoid overlapping subscriptions; Use hosted AI without managing provider API accounts; Chat across desktop and mobile devices; Document chat, knowledge bases and multi-model workflows
Official plans
Chatbox AI Free · USD 0; the managed free tier is separate from BYOK.
10,000 compute points per day; resets daily.
Document and image chat: supported · Knowledge base: not included · MCP: not included · Web search: supported · Work mode: trial only; ongoing entitlement unverified · AI images: 5; reset period unverified · Simultaneous devices: 5
Chatbox AI Lite · USD 3.99 monthly; USD 41.99 annually, displayed as USD 3.5/month (annual payment, not monthly collection; exact average = annual total / 12).
2,000,000 compute points per month; resets monthly.
Document and image chat: supported · Knowledge base: pending verification; current pricing does not include this feature; The current formal pricing page excludes knowledge base from Lite, while the older help table includes it. Current pricing takes priority; official reconciliation is pending. · MCP: pending verification; current pricing does not include this feature; The current formal pricing page excludes MCP services from Lite, while the older help table includes it. Current pricing takes priority; official reconciliation is pending. · Web search: supported · Work mode: supported · AI images: 25/month · Simultaneous devices: 5
Chatbox AI Pro · USD 19.99 monthly; USD 199.99 annually, displayed as USD 16.7/month (annual payment, not monthly collection; exact average = annual total / 12).
12,000,000 compute points per month; resets monthly.
Document and image chat: supported · Knowledge base: supported · MCP: supported · Web search: supported · Work mode: supported · AI images: 100/month · Simultaneous devices: 5
Compute points are the vendor's own unit, separate from tokens, CreditsPlan Credits and API dollar balances. Compare allowances only within the same unit and reset period.
The free allowance resets daily.
BYOK API charges are billed separately by the upstream provider. A downloaded client is not a free subscription or free API usage.
Supported models · 39
GPT 5.6 Luna · GPT 5.4 Mini · Claude Haiku 4.5 · Gemini 3 Flash · Qwen3.8 Max Preview · DeepSeek V4 Flash · DeepSeek V4 Pro · Chatbox AI 4 · GPT 6 Astra · GPT 6.1 Sol · GPT 6 Sol · GPT 5.6 Sol · GPT 5.6 Terra · GPT 5.5 · GPT 4o · Claude Fable 5.1 · Claude Fable 5 · Claude Opus 5.5 · Claude Opus 5 · Claude Opus 4.8 · Claude Opus 4.7 · Claude Opus 4.6 · Claude Sonnet 5.5 · Claude Sonnet 5 · Claude Sonnet 4.6 · Claude Sonnet 4.6 Thinking · Gemini 3.1 Pro · Gemini 2.5 Pro · Grok 4.7 · Grok 4.6 · Grok 4.5 · Grok 4 · MiMo V2.6 Pro UltraSpeed · Hy4 preview · Kimi K3 · Qwen3.8 Max · Qwen3.7 Max · Qwen3.7 Plus · Qwen3.6 Plus
A market label is not a guarantee of availability. Registration, payment and network conditions still apply.
Factory App
Desktop · Factory App
International
Mainland China support unverified
Factory Droid coding agents; rolling usage limits, BYOK and organization access conditions apply.
Coding
Agents
Source checked: 2026-10-12 Source verification does not certify the installer.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished. Approximately 5x Pro usage.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Approximately 5 times Pro usage; this does not establish fixed tokens or request counts.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished. Approximately 10x Pro usage.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Approximately 10 times Pro usage; this does not establish fixed tokens or request counts.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished. Pro limits apply to each seat.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Multipliers measure relative platform usage consumption, not API token prices. Billing countries, model service regions and per-plan model access are separate conditions.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished. Approximately 5x Pro usage.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Approximately 5 times Pro usage; this does not establish fixed tokens or request counts.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished. Approximately 10x Pro usage.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Approximately 10 times Pro usage; this does not establish fixed tokens or request counts.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished. Pro limits apply to each seat.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Multipliers measure relative platform usage consumption, not API token prices. Billing countries, model service regions and per-plan model access are separate conditions.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished. Approximately 5x Pro usage.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Approximately 5 times Pro usage; this does not establish fixed tokens or request counts.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished. Approximately 10x Pro usage.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Approximately 10 times Pro usage; this does not establish fixed tokens or request counts.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished. Pro limits apply to each seat.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Multipliers measure relative platform usage consumption, not API token prices. Billing countries, model service regions and per-plan model access are separate conditions.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished. Approximately 5x Pro usage.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Approximately 5 times Pro usage; this does not establish fixed tokens or request counts.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished. Approximately 10x Pro usage.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Approximately 10 times Pro usage; this does not establish fixed tokens or request counts.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Independent rolling 5-hour, 7-day and 30-day limits; exact quantities unpublished. Pro limits apply to each seat.
5-hour rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
7-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
30-day rolling usage window: the numeric allowance is unpublished; each window limits usage independently.
Standard Usage is consumed first; model usage and Droid Computers compute count toward it. It does not roll over.
Droid Core is a free open-weight model pool with separate rate limits. Exact allowance is unpublished; it is not unlimited.
Extra Usage is prepaid USD credit, minimum $10, with no expiry. The toggle stays on until disabled or exhausted; automatic top-up and exact consumption rates are unconfirmed.
Custom keys work in CLI/App, free up to the plan allowance, then charged according to the plan. Exact allowance and rate are unpublished; external provider fees are separate.
Missions share regular rolling limits, require Extra Usage to be enabled, and pause when a rate limit is reached.
The current request finishes; later requests are rate limited. Wait for reset or switch to Droid Core/Extra Usage through /limits.
Multipliers measure relative platform usage consumption, not API token prices. Billing countries, model service regions and per-plan model access are separate conditions.
VS Code：在扩展页搜索 Kilo Code，按官方当前指引选择 Install Pre-Release Version（官方说明该分发渠道稳定且推荐）；扩展 ID 为 kilocode.kilo-code。也可从官方链接进入 JetBrains Marketplace。安装后在 Kilo 内选择提供商并自行配置账号或 Key；助手不读取、保存或上传 Key。
Free software does not include paid inference. Configure primary, autocomplete and background models separately for free usage.
Credit top-ups and Pass purchase/renewal add a separate 5% processing fee without extra credits; Teams seats do not include credits.
Provider keys and existing Coding Plans are billed by their provider. Do not add inference already paid with credits a second time.
Cloud compute and inference are separate, billed per second; the $5 launch balance is not an extra charge.
Canceling renewals resets the streak. Annual Pass base credits continue to be issued monthly for the paid term.
Current Sep 30, 2026 terms limit purchased credits to one year or earlier account deletion. New users are covered immediately; existing users at the earliest acceptance, Desktop/artifact use or 30 days later. A billing help subsection still says never expire; check the applicable terms.
The FAQ says earliest-expiring credits first; terms say promotional, then monthly/purchased, then bonus. These conflict, so exact consumption-order estimates are disabled.
Organization Pass capacity is purchased per paid seat, pooled and adjusted with seat count. Personal subscriptions remain separate.
5% processing fee applies under the documented terms and is not added to the balance.
Code Review: USD 0.33/hour, billed per second; inference is separate.
Cloud Agent — Docker: USD 0.6/hour, billed per second; inference is separate.
Cloud Agent — Small: USD 0.6/hour, billed per second; inference is separate.
Cloud Agent — Standard: USD 1.2/hour, billed per second; inference is separate.
Count credit funding and uncovered usage once. BYOK provider bills are external costs. Model, usage and bonus eligibility are needed to estimate total cost.
Base prepaid model balance: USD 19; this is not unlimited calls.
Monthly: 10% in month 1; a one-time 50% welcome bonus in month 2 for first-time subscribers; variable tier/streak rewards from month 3 up to 40%, not fixed monthly. Annual: 50% each month. Unlock after monthly paid usage; unused bonus expires at the monthly service-period end.
Base prepaid model balance: USD 49; this is not unlimited calls.
Base prepaid model balance: USD 199; this is not unlimited calls.
Annual amount USD 2388 is derived as monthly price times 12; an independent annual quote and upfront charge are unconfirmed.
Supported models · 4
Anthropic: Claude Sonnet 4.6 · OpenAI: GPT-5.4 · Qwen: Qwen3 Coder Next · Auto Free
A market label is not a guarantee of availability. Registration, payment and network conditions still apply.
Kilo CLI
CLI · Kilo CLI
International
Mainland China support unverified
Open-source terminal coding agent using the kilo command; CLI, Desktop and IDE extensions are separate installations.
Coding
Agents
Source checked: 2026-10-12 Source verification does not certify the installer.
Free software does not include paid inference. Configure primary, autocomplete and background models separately for free usage.
Credit top-ups and Pass purchase/renewal add a separate 5% processing fee without extra credits; Teams seats do not include credits.
Provider keys and existing Coding Plans are billed by their provider. Do not add inference already paid with credits a second time.
Cloud compute and inference are separate, billed per second; the $5 launch balance is not an extra charge.
Canceling renewals resets the streak. Annual Pass base credits continue to be issued monthly for the paid term.
Current Sep 30, 2026 terms limit purchased credits to one year or earlier account deletion. New users are covered immediately; existing users at the earliest acceptance, Desktop/artifact use or 30 days later. A billing help subsection still says never expire; check the applicable terms.
The FAQ says earliest-expiring credits first; terms say promotional, then monthly/purchased, then bonus. These conflict, so exact consumption-order estimates are disabled.
Organization Pass capacity is purchased per paid seat, pooled and adjusted with seat count. Personal subscriptions remain separate.
5% processing fee applies under the documented terms and is not added to the balance.
Code Review: USD 0.33/hour, billed per second; inference is separate.
Cloud Agent — Docker: USD 0.6/hour, billed per second; inference is separate.
Cloud Agent — Small: USD 0.6/hour, billed per second; inference is separate.
Cloud Agent — Standard: USD 1.2/hour, billed per second; inference is separate.
Count credit funding and uncovered usage once. BYOK provider bills are external costs. Model, usage and bonus eligibility are needed to estimate total cost.
Base prepaid model balance: USD 19; this is not unlimited calls.
Monthly: 10% in month 1; a one-time 50% welcome bonus in month 2 for first-time subscribers; variable tier/streak rewards from month 3 up to 40%, not fixed monthly. Annual: 50% each month. Unlock after monthly paid usage; unused bonus expires at the monthly service-period end.
Base prepaid model balance: USD 49; this is not unlimited calls.
Base prepaid model balance: USD 199; this is not unlimited calls.
Annual amount USD 2388 is derived as monthly price times 12; an independent annual quote and upfront charge are unconfirmed.
Supported models · 4
Anthropic: Claude Sonnet 4.6 · OpenAI: GPT-5.4 · Qwen: Qwen3 Coder Next · Auto Free
A market label is not a guarantee of availability. Registration, payment and network conditions still apply.
Kilo Desktop
Desktop · Kilo Desktop
International
Mainland China support unverified
A free standalone desktop app in beta for code and data analysis; hosted inference is billed separately, with BYOK and local model options.
Coding
Agents
Source checked: 2026-10-12 Source verification does not certify the installer.
Free software does not include paid inference. Configure primary, autocomplete and background models separately for free usage.
Credit top-ups and Pass purchase/renewal add a separate 5% processing fee without extra credits; Teams seats do not include credits.
Provider keys and existing Coding Plans are billed by their provider. Do not add inference already paid with credits a second time.
Cloud compute and inference are separate, billed per second; the $5 launch balance is not an extra charge.
Canceling renewals resets the streak. Annual Pass base credits continue to be issued monthly for the paid term.
Current Sep 30, 2026 terms limit purchased credits to one year or earlier account deletion. New users are covered immediately; existing users at the earliest acceptance, Desktop/artifact use or 30 days later. A billing help subsection still says never expire; check the applicable terms.
The FAQ says earliest-expiring credits first; terms say promotional, then monthly/purchased, then bonus. These conflict, so exact consumption-order estimates are disabled.
Organization Pass capacity is purchased per paid seat, pooled and adjusted with seat count. Personal subscriptions remain separate.
5% processing fee applies under the documented terms and is not added to the balance.
Code Review: USD 0.33/hour, billed per second; inference is separate.
Cloud Agent — Docker: USD 0.6/hour, billed per second; inference is separate.
Cloud Agent — Small: USD 0.6/hour, billed per second; inference is separate.
Cloud Agent — Standard: USD 1.2/hour, billed per second; inference is separate.
Count credit funding and uncovered usage once. BYOK provider bills are external costs. Model, usage and bonus eligibility are needed to estimate total cost.
Base prepaid model balance: USD 19; this is not unlimited calls.
Monthly: 10% in month 1; a one-time 50% welcome bonus in month 2 for first-time subscribers; variable tier/streak rewards from month 3 up to 40%, not fixed monthly. Annual: 50% each month. Unlock after monthly paid usage; unused bonus expires at the monthly service-period end.
Base prepaid model balance: USD 49; this is not unlimited calls.
Base prepaid model balance: USD 199; this is not unlimited calls.
Annual amount USD 2388 is derived as monthly price times 12; an independent annual quote and upfront charge are unconfirmed.
Supported models · 4
Anthropic: Claude Sonnet 4.6 · OpenAI: GPT-5.4 · Qwen: Qwen3 Coder Next · Auto Free
A market label is not a guarantee of availability. Registration, payment and network conditions still apply.
