# TopxAI

> One OpenAI- and Anthropic-compatible API endpoint for Claude, GPT, Grok, GLM, Kimi, DeepSeek and TypeSafe's Jev at fixed USD prices, on prepaid credit; prompts, completions and files are relayed in memory and never stored.

TopxAI is the model API service run by TopXEA (https://topxea.com). Fourteen models on seven lines are reached through one base URL, https://ai.topxea.com/v1, with the SDK you already use: OpenAI Chat Completions and Responses, Anthropic Messages, image generation and edits, video generation, and /v1/systemone for Jev. Text models bill per million tokens, images per picture, video per second; the shared pool is 50% of the provider's list price, the official line 90%, DeepSeek 80% with its off-peak and peak hours, and an Auto-route key sends each request to the lowest-priced line that serves its model. The live price table is at https://ai.topxea.com/#models and as JSON at https://ai.topxea.com/api/pricing. GLM-5.3-Abliterated is a fine-tuned GLM 5.3 served from a private deployment, not the vendor's API.

## Models and prices

- [Every model's price](https://ai.topxea.com/pricing): one table, every route
- [LLM API price calculator](https://ai.topxea.com/tools/price-calculator): the cost of a workload on every model and route
- [Claude Fable 5.1](https://ai.topxea.com/pricing/claude-fable-5-1): claude-fable-5-1, $5 input / $25 output per 1M tokens; endpoints anthropic, openai
- [Claude Opus 5.5](https://ai.topxea.com/pricing/claude-opus-5-5): claude-opus-5-5, $2 input / $10 output per 1M tokens; endpoints anthropic, openai
- [Claude Sonnet 5.5](https://ai.topxea.com/pricing/claude-sonnet-5-5): claude-sonnet-5-5, $1 input / $5 output per 1M tokens; endpoints anthropic, openai
- [GPT-6 Astra](https://ai.topxea.com/pricing/gpt-6-astra): gpt-6-astra, $5 input / $25 output per 1M tokens; endpoints openai, openai-response
- [GPT-6.1 Sol](https://ai.topxea.com/pricing/gpt-6.1-sol): gpt-6.1-sol, $1 input / $5 output per 1M tokens; endpoints openai, openai-response
- [GPT Image 2.5 Flare](https://ai.topxea.com/pricing/gpt-image-2.5-flare): gpt-image-2.5-flare, from $0.03 per image; endpoints image-generation, openai, image-edit
- [GPT Image 2.5 Sunburst](https://ai.topxea.com/pricing/gpt-image-2.5-sunburst): gpt-image-2.5-sunburst, from $0.03 per image; endpoints image-generation, openai, image-edit
- [Grok 4.7](https://ai.topxea.com/pricing/grok-4.7): grok-4.7, $1 input / $3 output per 1M tokens; endpoints openai, openai-response
- [Grok Imagine Video 1.5](https://ai.topxea.com/pricing/grok-imagine-video-1.5): grok-imagine-video-1.5, from $0.05 per second; endpoints video-generation
- [GLM-5.3-Abliterated](https://ai.topxea.com/pricing/glm-5.3-abliterated): GLM-5.3-Abliterated, $4 input / $7 output per 1M tokens; endpoints openai
- [Kimi K3](https://ai.topxea.com/pricing/kimi-k3): kimi-k3, $2.4 input / $12 output per 1M tokens; endpoints openai
- [DeepSeek V4.1 Flash](https://ai.topxea.com/pricing/deepseek-flash): deepseek-flash, $0.12 input / $0.48 output per 1M tokens; endpoints openai
- [DeepSeek V4 Pro](https://ai.topxea.com/pricing/deepseek-v4-pro): deepseek-v4-pro, $0.528 input / $1.584 output per 1M tokens; endpoints openai
- [Jev 1.14.0](https://ai.topxea.com/pricing/jev-1.14.0): jev-1.14.0, $0.04494 per 1M input tokens, output free; endpoints systemone

## Documentation

- [Documentation index](https://ai.topxea.com/docs): every page in English and Chinese
- [What TopxAI is](https://ai.topxea.com/docs/what-topxai-is): One base URL, fourteen models on seven lines, prepaid USD credit, and nothing you send is stored.
- [Create an API key and choose a route](https://ai.topxea.com/docs/create-an-api-key-and-choose-a-route): The key drawer step by step, what each route means, and the three 403 refusals a key can hit.
- [Connect your SDK](https://ai.topxea.com/docs/connect-your-sdk): The base URL, every endpoint with its auth header, working examples for the OpenAI and Anthropic SDKs, and what each error status means.
- [API reference: every endpoint, with an OpenAPI specification](https://ai.topxea.com/docs/api-reference): Every endpoint TopxAI serves, what each accepts and returns, the two ways to send the key, and the OpenAPI 3.1 file at /openapi.json for API tools.
- [Claude Code with TopxAI: ANTHROPIC_BASE_URL and the key](https://ai.topxea.com/docs/use-topxai-with-claude-code): Point Claude Code at https://ai.topxea.com with two environment variables and pick claude-sonnet-5-5, claude-opus-5-5 or claude-fable-5-1 as its model.
- [Codex CLI custom model provider: model_providers in config.toml for TopxAI](https://ai.topxea.com/docs/use-topxai-with-codex-cli): Point Codex CLI at TopxAI with a [model_providers.topxai] table in ~/.codex/config.toml and run gpt-6.1-sol, gpt-6-astra or grok-4.7. An npm mirror is included.
- [Cursor with TopxAI: override the OpenAI base URL](https://ai.topxea.com/docs/use-topxai-with-cursor): In Cursor, paste a TopxAI key as the OpenAI key, turn on the base URL override with https://ai.topxea.com/v1 and add the model ids you want to chat with.
- [Cline and Roo Code with TopxAI: the OpenAI-compatible provider](https://ai.topxea.com/docs/use-topxai-with-cline-and-roo-code): Set the provider to OpenAI Compatible with base URL https://ai.topxea.com/v1, a TopxAI key and a model id; or the Anthropic provider with a custom base URL.
- [Continue with TopxAI: models in config.yaml](https://ai.topxea.com/docs/use-topxai-with-continue): Add TopxAI models to ~/.continue/config.yaml with provider openai or anthropic and apiBase, then use them for chat, edit and agent mode in VS Code or JetBrains.
- [Zed with TopxAI: OpenAI- and Anthropic-compatible providers](https://ai.topxea.com/docs/use-topxai-with-zed): Add TopxAI to Zed as an openai_compatible provider for GPT, Grok, Kimi and GLM, and as an anthropic_compatible provider for Claude, in settings.json.
- [Aider with TopxAI: OPENAI_API_BASE and the model name](https://ai.topxea.com/docs/use-topxai-with-aider): Run aider against TopxAI with OPENAI_API_BASE=https://ai.topxea.com/v1 and --model openai/claude-sonnet-5-5, or the Anthropic variables for the Claude line.
- [opencode with TopxAI: a custom provider in opencode.json](https://ai.topxea.com/docs/use-topxai-with-opencode): Declare TopxAI in opencode.json with @ai-sdk/openai-compatible, the base URL and the model ids, then pick them with /models.
- [Open WebUI with TopxAI: one OpenAI connection, every model](https://ai.topxea.com/docs/use-topxai-with-open-webui): Add https://ai.topxea.com/v1 as an OpenAI API connection in Open WebUI; the model list fills from /v1/models and every text model is ready to chat.
- [LobeChat with TopxAI: the OpenAI proxy address](https://ai.topxea.com/docs/use-topxai-with-lobechat): Set OpenAI’s API proxy address to https://ai.topxea.com/v1 in LobeChat, or OPENAI_PROXY_URL for a deployment, and fetch the model list.
- [Cherry Studio with TopxAI: first chat and endpoint setup](https://ai.topxea.com/docs/use-topxai-with-cherry-studio): Connect a route-scoped TopxAI key to Cherry Studio, choose the right endpoint, and diagnose the first request without guessing URL suffixes.
- [ChatBox with TopxAI: a custom provider](https://ai.topxea.com/docs/use-topxai-with-chatbox): In ChatBox, add a custom provider in OpenAI API compatible mode with host https://ai.topxea.com and path /v1/chat/completions, then add the model ids.
- [LangChain with TopxAI: ChatTopxAI from the langchain-topxai package](https://ai.topxea.com/docs/use-topxai-with-langchain): Call TopxAI chat models from LangChain in Python with ChatTopxAI from langchain-topxai. ChatOpenAI with a base URL still works, also for the Responses API.
- [Vercel AI SDK with TopxAI: createOpenAI, createAnthropic and openai-compatible](https://ai.topxea.com/docs/use-topxai-with-vercel-ai-sdk): Create an AI SDK provider with baseURL https://ai.topxea.com/v1 and a TopxAI key, then use generateText or streamText with any catalogue model.
- [LiteLLM with TopxAI: proxy config and the Python SDK](https://ai.topxea.com/docs/use-topxai-with-litellm): Route LiteLLM to TopxAI with model openai/<id> and api_base https://ai.topxea.com/v1, in the proxy’s model_list or a completion() call.
- [Dify with TopxAI: the OpenAI-API-compatible provider](https://ai.topxea.com/docs/use-topxai-with-dify): Add each TopxAI model to Dify through the OpenAI-API-compatible provider with endpoint https://ai.topxea.com/v1, your key and the model id.
- [n8n with TopxAI: an OpenAI credential with a base URL](https://ai.topxea.com/docs/use-topxai-with-n8n): Create an OpenAI credential in n8n with base URL https://ai.topxea.com/v1 and a TopxAI key, then use any catalogue model in the Chat Model and AI Agent nodes.
- [llm CLI with TopxAI: extra-openai-models.yaml](https://ai.topxea.com/docs/use-topxai-with-llm-cli): Register TopxAI models in llm’s extra-openai-models.yaml with api_base https://ai.topxea.com/v1, store the key with llm keys set, and prompt from the shell.
- [Shared pool, official line and auto route](https://ai.topxea.com/docs/shared-pool-official-line-and-auto-route): Same models, different routing and price: what each API-key route means and how Auto route picks one for you.
- [How requests are billed](https://ai.topxea.com/docs/how-requests-are-billed): Billing units, long-context tiers, the up-front reservation and what happens to it when a request fails.
- [Reading your usage log](https://ai.topxea.com/docs/reading-your-usage-log): What each row of the usage log shows, how to filter it, and what is never recorded.
- [Jev (TypeSafe System One)](https://ai.topxea.com/docs/jev-typesafe-system-one): What Jev 1.14 returns, how to call POST /v1/systemone with text or images, Box questions, the one-line setup for a TypeSafe key, and the price.
- [GLM-5.3-Abliterated](https://ai.topxea.com/docs/glm-5-3-abliterated): A fine-tuned GLM 5.3 on a private deployment: text only, 1M-token context, three reasoning modes, priced above list, and covered by a compliance notice.
- [Image generation](https://ai.topxea.com/docs/image-generation): gpt-image-2.5-sunburst and -flare: the two endpoints, the 1K and 2K per-image prices, how the size tier is chosen, and where the image link comes from.
- [Video generation](https://ai.topxea.com/docs/video-generation): grok-imagine-video-1.5: submit a job, poll it, per-second prices by resolution, refunds on failure, and how forgotten jobs are refunded after 24 hours.
- [Top up with a card](https://ai.topxea.com/docs/top-up-with-a-card): Add USD credit with a card, Apple Pay, Google Pay or a local e-wallet; the checkout runs on topxea.com through Waffo.
- [Pay with crypto](https://ai.topxea.com/docs/pay-with-crypto): Top up with USDT or USDC through a NOWPayments invoice: what to send, how long confirmation takes, and what happens if you send too little.
- [Gift cards and credit codes](https://ai.topxea.com/docs/gift-cards-and-credit-codes): Buy a single-use credit code, redeem one on any account, and what the User Agreement says about refunds.
- [Common error responses](https://ai.topxea.com/docs/common-error-responses): What each HTTP status and error code from the gateway means, and what to change before you retry.
- [The model works in curl but not in my tool](https://ai.topxea.com/docs/the-model-works-in-curl-but-not-in-my-tool): Seven places where a tool differs from the working curl: base URL, auth header, model id, key route and limits, IP whitelist, GLM and Jev endpoints, timeouts.
- [Sign-in and account](https://ai.topxea.com/docs/sign-in-and-account): How sign-in is shared with topxea.com, how sessions, devices and TopxAI Desktop sign-ins work, the regional service notice, and how to reach support.
- [TopxAI vs OpenRouter: what is the same and what differs](https://ai.topxea.com/docs/topxai-vs-openrouter): Compare current model and route prices: TopxAI shared routes cost 37.5% to 50% less than OpenRouter standard routes, and Kimi hosts can cost less.
- [TopxAI vs the providers' own APIs: price, retention and what stays direct](https://ai.topxea.com/docs/topxai-vs-the-providers-own-apis): What changes when Claude, GPT, Grok or Kimi go through TopxAI: the price per token, one key and one balance, zero retention, and what stays with the provider.
- [TopxAI vs Vercel AI Gateway: list price with no markup, or half of it](https://ai.topxea.com/docs/topxai-vs-vercel-ai-gateway): Vercel AI Gateway charges the provider list price with no markup; TopxAI shared routes charge 50% of it for Claude, GPT and Grok. What else differs.
- [What TopxAI stores and what it never stores](https://ai.topxea.com/docs/what-topxai-stores-and-what-it-never-stores): Request and response content passes through memory only; here is what the usage log, account and payment records keep, and for how long.
- [Acceptable use and abuse reports](https://ai.topxea.com/docs/acceptable-use-and-abuse-reports): The six prohibited content categories, how enforcement works, why GLM's private deployment is no exception, and how to report abuse.
- [Full documentation text](https://ai.topxea.com/llms-full.txt): all pages in one file

## Blog

- [Jev 1.14 on TopxAI: images, Box questions and what they cost](https://ai.topxea.com/blog/jev-1-14-images-and-box): Jev 1.14 replaces 1.13 at the same price and can look at images. Our measurements of Box accuracy, image token counts and the real size limits.
- [Moving to Claude Sonnet 5.5 on TopxAI](https://ai.topxea.com/blog/claude-sonnet-5-5): Update the Sonnet model ID, keep the same token prices, and check thinking and tool settings when moving your client to Claude Sonnet 5.5.
- [Moving to GPT-6.1 Sol on TopxAI](https://ai.topxea.com/blog/gpt-6-1-sol): The new model ID, lower cache-read prices, reasoning settings and a Responses API example for the GPT-6.1 Sol upgrade.
- [AI coding agents are moving the bottleneck to verification](https://ai.topxea.com/blog/ai-agents-verification-bottleneck): What recent research says about agent productivity, why more output can slow delivery, and a practical evaluation worksheet for small engineering teams.
- [An AI API timeout is an unknown outcome, not a failed action](https://ai.topxea.com/blog/ai-api-timeouts-retries-unknown-outcomes): How to separate generation retries from tool execution, recover after an interrupted stream, and test duplicate effects before shipping an agent.
- [Why I paused an AI project: the cache bill missing from the architecture](https://ai.topxea.com/blog/jevis-mvp-prompt-cache-reality): Jevis had a detailed design for frontier-model planning and cheaper execution. Prompt-cache economics exposed the hypothesis we still had to test.
- [TypeSafe Jev: a System One model that returns judgments, not text](https://ai.topxea.com/blog/typesafe-jev-system-one): What Jev (jev-1.14.0) is, how its noul, choice, score and Box questions work, where it is weak, what it costs, and how to call it through TopxAI.

## Changelog

- [Changelog](https://ai.topxea.com/changelog): every dated change; RSS at https://ai.topxea.com/changelog/feed.xml
- [Jev 1.14 replaces Jev 1.13, with images and Box questions](https://ai.topxea.com/changelog#jev-1-14-images-and-box): Use jev-1.14.0 on /v1/systemone at the same price. Questions can now look at PNG and JPEG images, and the new Box type finds where something is in a picture. (2026-10-16)
- [Claude Sonnet 5.5 replaces Sonnet 5 at the same prices](https://ai.topxea.com/changelog#claude-sonnet-5-5): Use claude-sonnet-5-5 on both Claude routes. Input, output and cache prices stay the same; update saved model IDs and restricted keys. (2026-10-02)
- [GPT-6.1 Sol replaces GPT-6 Sol](https://ai.topxea.com/changelog#gpt-6-1-sol): Upgrade to gpt-6.1-sol: shared routes remain $1 input and $5 output per million tokens, with cache reads reduced to $0.05. (2026-09-30)
- [GLM-5.3-Abliterated publishes its reasoning modes and 1M-token context](https://ai.topxea.com/changelog#glm-5-3-abliterated-reasoning-modes): The price feed now lists the low, high and max reasoning modes and the 1,000,000-token window, and the Z.ai-style thinking switch no longer fails with a 400. (2026-09-27)
- [Claude Fable 5 is retired; use Fable 5.1](https://ai.topxea.com/changelog#claude-fable-5-retired): claude-fable-5 is no longer offered. Claude Fable 5.1 stays at the same input and output prices with a lower cache-read price. (2026-09-24)
- [Old model IDs claude-opus-5, gpt-5.6-sol and grok-4.6 are retired](https://ai.topxea.com/changelog#old-model-ids-retired): The three old IDs no longer forward to their successors. Use claude-opus-5-5, gpt-6-sol and grok-4.7; prices do not change. (2026-09-24)
- [Claude Opus 5.5 replaces Opus 5 with lower token prices](https://ai.topxea.com/changelog#claude-opus-5-5): Opus 5.5 on both Claude routes at 50% and 90% of the Anthropic list price, including cache reads and writes. Existing Opus 5 API calls remain compatible. (2026-09-23)
- [GPT-6 Sol is available at half the OpenAI Standard price](https://ai.topxea.com/changelog#gpt-6-sol): GPT-6 Sol replaces GPT-5.6 Sol, with shared routes at 50% and official routes at 90% of Standard list rates. Existing model IDs and keys remain compatible. (2026-09-23)
- [Grok 4.7 is available at half the xAI Standard price](https://ai.topxea.com/changelog#grok-4-7): Grok 4.7 replaces Grok 4.6 at 50% and 90% of xAI prices, with discounted cache reads and compatibility for existing model IDs and keys. (2026-09-23)
- [Long-context prices for GPT-6 Sol and Grok 4.7](https://ai.topxea.com/changelog#long-context-prices-for-gpt-6-sol-and-grok-4-7): GPT-6 Sol above 272K input tokens and Grok 4.7 from 200K bill the whole request at long-context rates, as on OpenAI and xAI. (2026-09-23)
- [Codex CLI downloads and a one-command installer](https://ai.topxea.com/changelog#codex-downloads-and-installer): A Downloads page lists the official Codex CLI packages through an npm mirror; one command installs Codex and connects it to TopxAI on macOS, Linux or Windows. (2026-09-22)
- [DeepSeek V4.1 Flash and V4 Pro join the catalogue, with off-peak and peak hours](https://ai.topxea.com/changelog#deepseek-joins-the-catalogue): deepseek-flash and deepseek-v4-pro on one official line at 80% of list in both provider tiers; every usage-log row names the tier it was billed at. (2026-09-22)
- [Auto-route keys, the blog, three more languages and the help messenger](https://ai.topxea.com/changelog#auto-route-keys-blog-and-five-languages): One auto-route key for every model, a blog, Japanese, Russian and Spanish interfaces, the shared help messenger and the acceptable use policy. (2026-09-18)
- [Card payments on topxea.com and gift cards](https://ai.topxea.com/changelog#card-payments-and-gift-cards): Card top-ups on topxea.com, and gift cards that can be bought on either site and redeemed for TopxAI credit. (2026-09-18)
- [GLM-5.3-Abliterated, Kimi K3 and Jev join the catalogue](https://ai.topxea.com/changelog#glm-kimi-and-jev-join-the-catalogue): GLM-5.3-Abliterated on a private deployment, Kimi K3 at 80% of list and TypeSafe Jev through /v1/systemone with a one-command setup. (2026-09-18)
- [Image and video generation, long-context tiers](https://ai.topxea.com/changelog#image-and-video-generation-and-long-context-tiers): GPT Image 2.5 per image, Grok Imagine Video 1.5 per second, and whole-request long-context tiers for OpenAI and Grok models. (2026-09-18)
- [Credit codes, shared sign-in with topxea.com and the regional notice](https://ai.topxea.com/changelog#credit-codes-shared-sign-in-and-regional-notice): Redeem credit codes on the wallet page, sign in with the same account as topxea.com, and a notice for regions the upstreams do not serve. (2026-09-12)
- [TopxAI is live: Claude, OpenAI and Grok through one endpoint](https://ai.topxea.com/changelog#topxai-is-live): The first release: Claude, OpenAI and Grok models on shared pool and official line routes through one OpenAI- and Anthropic-compatible endpoint. (2026-09-11)

## Labs

- [Labs](https://ai.topxea.com/labs): Tools and kits from our own work: the BinanceGrid EA for MetaTrader 5 (sold on TopXEA), and the free TopxAI Equity Research agent skill, TopX HUD plugin for Claude Code and TopX Terminal design system.
- [BinanceGrid: a Binance USD-M futures grid EA for MetaTrader 5](https://ai.topxea.com/labs/binancegrid): A grid Expert Advisor that trades Binance USD-M perpetual futures (HYPE, LIT, ADA) from MetaTrader 5, with Binance data on MT5 custom symbols, paper or live execution and an on-chart dashboard. Backtest +18.9% with a 4.8% max equity drawdown. $599 with the full MQL5 source code, sold on TopXEA.
- [TopxAI Equity Research: an open-source agent skill for equity research](https://ai.topxea.com/labs/topxai-equity-research): A free, open-source agent skill and Claude Code plugin that runs a research desk: filings-first data from SEC EDGAR, falsifiable calls, a bull/bear debate before every rating, adversarial audits and a ledger that settles every call.
- [TopX HUD: context, limits and one-click compact above the Claude Code prompt](https://ai.topxea.com/labs/topx-hud): A free Claude Code plugin that shows context use, the 5-hour and 7-day limits and whether a turn is running above the prompt, with a Compact button and a /topx pane of recent turns.
- [TopX Terminal design system: tokens, CSS and components](https://ai.topxea.com/labs/topx-terminal): The design system of a quant trading terminal: neutral grey, dense tables and colour with one meaning each. Light and dark tokens, CSS and component previews, free to reuse.

## Policies

- [Hosted MCP servers](https://ai.topxea.com/mcp): Use Claude, GPT, Grok, GLM, Kimi, DeepSeek, Jev, GPT Image and Grok video from inside Claude Code, Cursor or any MCP client: the same API key, the same prices, plus a small fee per tool call.
- [TopxAI Desktop downloads](https://ai.topxea.com/downloads): Download TopxAI Desktop v1.52.4 for Apple silicon Macs running macOS 15 or later. Get the DMG, release notes and first-launch instructions. Windows is coming soon.
- [About TopxAI · One API for Claude, GPT, Grok, GLM, Kimi, DeepSeek, Jev](https://ai.topxea.com/about): TopXEA's model API: 14 models from Claude, OpenAI, Grok, GLM, Kimi, DeepSeek and Jev on one OpenAI- and Anthropic-compatible endpoint, fixed USD prices, no stored prompts.
- [Privacy Policy](https://ai.topxea.com/privacy-policy): What TopxAI keeps and what it never stores: account, API key and payment records are retained for accounting; prompts, completions and uploaded files are relayed in memory and never persisted.
- [User Agreement](https://ai.topxea.com/user-agreement): Terms for using the TopxAI API relay: prepaid USD credit, usage-based charges, acceptable use, refunds, liability and account closure.
- [AI Acceptable Use Policy](https://ai.topxea.com/acceptable-use-policy): Content standards for traffic relayed through TopxAI to Claude, OpenAI, Grok, GLM, Kimi and Jev: prohibited sexual, violent, hateful, child-endangering, deepfake and infringing content, graduated enforcement, and how to report abuse.

## Optional

- [OpenAPI 3.1 specification](https://ai.topxea.com/openapi.json): every endpoint with its request and response schemas
- [Documentation mirror on GitHub](https://github.com/UnbelievableT/topxai-docs): the same pages as Markdown
- [TopXEA](https://topxea.com): the operator; MetaTrader expert advisors and research
- [X @topxea](https://x.com/topxea): announcements
- [Telegram @topxea](https://t.me/topxea): announcements
- Support: support@topxea.com