# OmniaKey

> Unified AI API for LLM, image, video and music models. One prepaid key connects agents and apps to multiple providers with usage-based billing. Requests use the model ID you select, without silent substitution.

OmniaKey is a unified AI API platform. Text clients choose the documented OpenAI-compatible, Anthropic-native or Gemini-native API; image, video and music use asynchronous media tasks. Model availability, supported protocols, billing units and current prices are maintained in the catalog and docs.

## Endpoints

- OpenAI-compatible: `https://api.omniakey.com/v1`
- Anthropic-compatible: `https://api.omniakey.com` (SDK appends `/v1/messages`)
- Gemini-compatible: `https://api.omniakey.com/v1beta`

## Public API scope

- OpenAI-compatible Chat Completions: `POST /v1/chat/completions`
- OpenAI-compatible Responses: `POST /v1/responses`
- OpenAI-compatible Responses compaction: `POST /v1/responses/compact`
- OpenAI-compatible model discovery: `GET /v1/models`
- Anthropic Messages: `POST /v1/messages`
- Gemini Generate Content: `POST /v1beta/models/{model}:generateContent`
- Gemini model discovery: `GET /v1beta/models`
- Media tasks (image, video, music): `POST /v1/media/tasks` (requires an `Idempotency-Key` header), poll `GET /v1/media/tasks/{id}`, cancel `POST /v1/media/tasks/{id}/cancel`; finished assets are returned as signed URLs

Do not state that OmniaKey publicly supports embeddings, rerank, files, fine-tuning, or batches unless those capabilities appear in the docs and dashboard catalog.

## Canonical positioning

- Category: unified AI API platform — one prepaid key for LLM, image, video and music models, with text protocols and asynchronous media tasks
- Current public catalog: curated Claude, GPT, Gemini, Grok, Qwen, DeepSeek, GLM, Kimi, MiniMax, MiMo, Hunyuan, and Jev models, plus image, video, and music models served as asynchronous media tasks
- Primary users: developers using Claude Code, Cursor, Codex, Cline, aider, OpenAI-compatible tools, Anthropic-compatible tools, Gemini-compatible tools, and model-agnostic agent platforms
- Billing: prepaid balance, per-token and per-request usage (media jobs per call or per second of output), no monthly subscription
- Model policy: no silent fallback, no quantization, no distillation, no model substitution
- Data policy: prompt and response bodies are not stored by default; billing metadata is retained for usage analytics
- Independent monitoring: OkkMax, a third-party monitor of API relay stations, probes OmniaKey's availability, latency and Claude model authenticity: https://www.okkmax.com/station/omniakey

## For agents

- [Full reference](https://omniakey.com/llms-full.txt): about, setup, FAQ and every price in one file
- [Pricing](https://omniakey.com/pricing.md): every live route with OmniaKey and official prices, generated from the catalog at deploy
- [About](https://omniakey.com/about.md): what OmniaKey is, model and data policy, contact, and independent monitoring
- [Docs index](https://docs.omniakey.com/llms.txt): every guide and API page

## Models

- Claude (Anthropic): claude-fable-5, claude-fable-5-1, 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-haiku-5-5, claude-haiku-4-5
- GPT (OpenAI): gpt-6.1-sol, gpt-6-astra, gpt-6-sol, gpt-6-luna, gpt-5.6-sol, gpt-5.6-terra, gpt-5.6-luna, gpt-5.5
- Gemini (Google): gemini-3.1-pro-preview, gemini-4-argon, gemini-2.5-pro, gemini-3.8-flash, gemini-3.7-flash, gemini-3.6-flash, gemini-3.5-flash-lite, gemini-2.5-flash, gemini-2.5-flash-lite
- Grok (xAI): grok-4.7, grok-4.6, grok-4.5
- Qwen (Alibaba Cloud): qwen3.8-max, qwen3.8-flash, qwen3.7-max, qwen3.7-plus, qwen3.7-flash
- DeepSeek: deepseek-flash, deepseek-v4-flash, deepseek-v4-flash-vision-exp, deepseek-v4-pro
- GLM (Z.ai): glm-5.3, glm-5.3-flash, glm-5.2
- Kimi (Moonshot AI): kimi-k3, kimi-k2.7-code, kimi-k2.6
- MiniMax: MiniMax-M3, MiniMax-M2.7, MiniMax-M2.7-highspeed
- MiMo (Xiaomi): mimo-v2.5, mimo-v2.5-pro
- Hunyuan (Tencent): hy3, hy4-preview
- Jev (TypeSafe AI): jev-latest
- Image (OpenAI, Google, Alibaba, ByteDance, xAI): gpt-image-2, gpt-image-2.5-flare, gpt-image-2.5-sunburst, gemini-3.1-flash-image, gemini-3.1-flash-image-preview, gemini-nano-banana-2.1, gemini-3.1-flash-lite-image, gemini-2.5-flash-image, gemini-3-pro-image, gemini-3-pro-image-preview, qwen-image-3, qwen-image-3-pro, wan2.7-image-pro, Dola-Seedream-5.0-pro, Dola-Seedream-5.0-lite, grok-imagine-image, grok-imagine-image-quality, grok-imagine-image-2.0
- Video (xAI, Alibaba, ByteDance, MiniMax, Kling): grok-imagine-video-1.5, happyhorse-1.1-t2v, happyhorse-1.1-i2v, happyhorse-1.1-r2v, happyhorse-1.0-t2v, happyhorse-1.0-i2v, happyhorse-1.0-r2v, wan3.0-video, wan3.0-video-prime, dreamina-seedance-2-5-260628, dreamina-seedance-2-0-260128, dreamina-seedance-2-0-fast-260128, dreamina-seedance-2-0-mini-260615, minimax-h3, kling-3.0
- Music (Suno): suno-v4.5

## FAQ

### What is OmniaKey?

OmniaKey is a unified AI API platform for agents and applications. One prepaid key covers language models and asynchronous image, video and music generation. Compare model capabilities, supported protocols and published prices before choosing a route.

### Does OmniaKey replace model providers?

No. OmniaKey routes to Anthropic, OpenAI, Google, xAI, Alibaba Cloud, DeepSeek, Z.ai, Moonshot AI, MiniMax, Xiaomi, Tencent, and TypeSafe AI models, and to ByteDance, Kling, and Suno media models. The requested provider and model are what run.

### Which coding agents work with OmniaKey?

Claude Code, Cursor, Codex, Cline, aider, and other OpenAI-, Anthropic-, or Gemini-compatible tools can use OmniaKey.

### What are the correct tool base URLs?

Claude Code uses `ANTHROPIC_BASE_URL=https://api.omniakey.com` and `ANTHROPIC_AUTH_TOKEN=<key>`. OpenAI-compatible tools use `https://api.omniakey.com/v1`. Gemini-native HTTP uses `https://api.omniakey.com/v1beta`.

### What is the current Codex CLI provider shape?

Codex custom providers should use `wire_api="responses"`, `base_url="https://api.omniakey.com/v1"`, and `env_key="OMNIAKEY_API_KEY"`. The API key is read from `OMNIAKEY_API_KEY` in the shell that launches Codex. Do not combine `env_key` with `requires_openai_auth=true`; Codex ignores `env_key` in that mode.

### How should SDKs connect?

Use `https://api.omniakey.com/v1` for OpenAI-compatible clients, `https://api.omniakey.com` for Anthropic-compatible clients, and `https://api.omniakey.com/v1beta` for Gemini-compatible clients.

### How does pricing work?

Users top up a prepaid balance and pay per token; media jobs are billed per call or per second of output. Current line-by-line pricing is maintained on the pricing page and in the dashboard.

## Site

- [Home](https://omniakey.com): product overview, protocols, and the model line-up
- [Models catalog](https://omniakey.com/models): every model live on the gateway, with context windows and rates
- [Pricing](https://omniakey.com/pricing): per-token and per-job pricing against official provider list prices
- [Blog](https://omniakey.com/blog): model reviews, pricing breakdowns, and coding-agent setup guides

## Docs

- [Docs home](https://docs.omniakey.com): full product documentation
- [Quick start](https://docs.omniakey.com/en/quick-start): create a key and send the first request
- [API reference](https://docs.omniakey.com/en/api): reference index for all three protocols
- [API overview](https://docs.omniakey.com/en/api/overview): how the three protocols map onto one key
- [Authentication](https://docs.omniakey.com/en/api/authentication): key formats and auth headers per protocol
- [OpenAI-compatible API](https://docs.omniakey.com/en/api/openai-compatible): Chat Completions, Responses, and model discovery
- [Anthropic-native API](https://docs.omniakey.com/en/api/anthropic-native): the Messages endpoint
- [Gemini-native API](https://docs.omniakey.com/en/api/gemini-native): generateContent and model discovery
- [Errors and rate limits](https://docs.omniakey.com/en/api/errors): error shapes, status codes, and retry guidance
- [Models documentation](https://docs.omniakey.com/en/models): model ids, aliases, and capability notes
- [Media tasks](https://docs.omniakey.com/en/media/overview): create, poll, and cancel asynchronous image, video, and music jobs
- [Image generation](https://docs.omniakey.com/en/media/image): GPT Image, Nano Banana, Qwen, Wan, Seedream, and Grok Imagine models
- [Video generation](https://docs.omniakey.com/en/media/video): Grok Imagine Video, HappyHorse, Wan 3.0, Seedance, MiniMax H3, and Kling 3.0
- [Audio generation](https://docs.omniakey.com/en/media/audio): Suno v4.5 songs from a description or your own lyrics

## Coding agent guides

- [Coding agents overview](https://docs.omniakey.com/en/coding): which agents work over which protocol
- [Claude Code](https://docs.omniakey.com/en/guides/claude-code-api-key): base URL and auth token setup
- [Cursor](https://docs.omniakey.com/en/guides/cursor-api-key): custom API key and base URL setup
- [Codex CLI](https://docs.omniakey.com/en/guides/codex-cli): custom provider block for the Responses wire API
- [Cline](https://docs.omniakey.com/en/guides/cline): OpenAI-compatible provider setup
- [aider](https://docs.omniakey.com/en/guides/aider): model and base URL flags
- [OpenAI SDK](https://docs.omniakey.com/en/guides/openai-sdk): drop-in base URL change for Python and Node.js

## Blog

- [Muse Wonder guide](https://omniakey.com/blog/muse-wonder): Wonder versus Full access, direct Meta API token prices, cached-input and reasoning cost calculations, Muse Code subscription boundaries, independent benchmarks, and a documented request; check the current catalog for OmniaKey availability.

- [Kling 3.0 vs Seedance 2.5](https://omniakey.com/blog/kling-3-0-vs-seedance-2-5): per-generation matrix pricing against per-second resolution rates, live OmniaKey prices and wall-clock observations, the Kling duration cap and missing reference operation, Seedance 2.0 Fast and Mini tiers, and worked per-clip costs without a quality verdict.

- [Ideogram 4.5 review](https://omniakey.com/blog/ideogram-4-5-review): precise editing, attributed official samples, typography checks, 4.0 API migration, request-based pricing, ComfyUI and commercial-use terms; no new generation benchmark or announcement of OmniaKey availability.

- [Universal Modder guide](https://omniakey.com/blog/universal-modder): Original toolkit, Claude Code installation, version-specific game examples, separate fal costs, and a first-mod verification and rollback plan.
- [Anthropic OSS Scanner](https://omniakey.com/blog/anthropic-oss-scanner): free security scans for eligible open-source projects, enrollment configuration, offline audit preparation, unreviewed reports and disclosure policy.
- [Haiku 5.5 vs Sonnet 5](https://omniakey.com/blog/claude-haiku-5-5-vs-sonnet-5): direct API prices across the 100K prompt threshold, coding task selection, speed evidence, effort defaults and cost per accepted task
- [Haiku 5.5 vs Luna 6](https://omniakey.com/blog/haiku-5-5-vs-luna-6): Claude Haiku 5.5 versus GPT-6 Luna, official and OmniaKey prices, 100K/272K thresholds, cached-input calculations, coding evaluation and API differences; a sourced comparison without a new benchmark
- [Claude Haiku 5.5 API pricing](https://omniakey.com/blog/claude-haiku-5-5-api-pricing): 100K prompt tiers, half-price arithmetic, cache and Batch costs, reproducible bills, and the boundary between Anthropic list rates and gateway quotes
- [Claude Haiku 5.5 review](https://omniakey.com/blog/claude-haiku-5-5-review): official benchmarks, $0.10/$0.50 direct API pricing, adaptive thinking, tokenizer changes, and migration from Haiku 4.5; check the live catalog for OmniaKey availability
- [Claude Opus 5.5 vs Sonnet 5.5](https://omniakey.com/blog/claude-opus-5-5-vs-sonnet-5-5): exact-version coding comparison, attributed benchmark limitations, cache-inclusive cost calculations, effort defaults, migration constraints and a reproducible evaluation method

- [Vidu Q4 Preview guide](https://omniakey.com/blog/vidu-q4-preview): official image and reference video limits, API credit pricing, request example, ten-second shot plan and cost per accepted clip
- [Opus 5.5 video guide](https://omniakey.com/blog/claude-opus-5-5-video): attributed examples, original storyboard prompt, Claude Code and HyperFrames workflow, MP4 export, rendering costs and the boundary between code animation and native video generation
- [Claude Sonnet 5.5 review](https://omniakey.com/blog/claude-sonnet-5-5-review): Sonnet 5.5 pricing and cost, direct API token and cache rates, task calculations, benchmark evidence, and migration requirements
- [Codex usage limit reached](https://omniakey.com/blog/codex-usage-limit-reached): reset times, weekly limits, extra ChatGPT credits, and separately billed local API options
- [Meta Muse Agent review](https://omniakey.com/blog/meta-muse-agent-review): personal-agent capabilities, invite-code rewards, September updates, attributed media experiences, and privacy controls

- [Claude Opus 5.5 review](https://omniakey.com/blog/claude-opus-5-5-review): official release, $4/$20 direct API pricing, Opus 5 comparison, benchmark caveats, and breaking API migration changes; check the model catalog for OmniaKey availability
- [Grok 4.7 review](https://omniakey.com/blog/grok-4-7-review): release verdict, Grok 4.6 comparison, official and independent benchmarks, API pricing, 500K context, and migration checks
- [Qwen-Image-2.1 review](https://omniakey.com/blog/qwen-image-2-1-review): native 2K and RGBA output, editing, benchmarks, VRAM, ComfyUI setup, and the non-commercial research license
- [Gemini 4 release date](https://omniakey.com/blog/gemini-4-release-date): what Google confirmed, what remains unannounced, how to verify API availability, and which current model to use
- [Gemini 4 Argon review](https://omniakey.com/blog/gemini-4-argon-review): launch evidence, benchmark interpretation, staged access, provider pricing, and API availability checks
- [Gemini 3.8 Flash API pricing](https://omniakey.com/blog/gemini-3-8-flash-api-pricing): Google and OmniaKey rates, the January 2027 change, and retry-aware cost per accepted agent task
- [Jev model explained](https://omniakey.com/blog/jev-model-explained): Jev API pricing, typed decisions, and immediate OmniaKey access without a TypeSafe waitlist
- [GLM-5.3-FlashX review](https://omniakey.com/blog/glm-5-3-flashx-review): 200 token/s evidence, direct API pricing, regular Flash baseline, cost math, and latency-based routing
- [Cherry Studio API setup](https://omniakey.com/blog/cherry-studio-api-setup): connect GPT and third-party providers, add model IDs, and troubleshoot API errors
- [Claude Code setup](https://omniakey.com/blog/claude-code-setup): configure the Anthropic-native endpoint, API key, and exact Claude model
- [OpenClaw setup](https://omniakey.com/blog/openclaw-setup): configure an OpenAI-compatible provider and model allowlist
- [Hermes Agent setup](https://omniakey.com/blog/hermes-setup): connect a custom OpenAI-compatible endpoint and switch models
- [Best LLM for coding agents](https://omniakey.com/blog/best-llm-coding-agents): Claude, GPT, and Gemini trade-offs, accepted-task cost, and model routing
- [Wan 3.0 vs Wan 2.7](https://omniakey.com/blog/wan-3-0-vs-wan-2-7): verified duration, input, editing, pricing, live availability, and migration boundaries for Alibaba's hosted video APIs
- [GPT-6 Astra vs GPT-5.6 Sol](https://omniakey.com/blog/gpt-6-astra-vs-gpt-5-6-sol): direct API pricing, migration differences, break-even math, and workload routing
- [GPT-6.1 Sol review](https://omniakey.com/blog/gpt-6-1-sol-review): attributed same-effort coding benchmarks, direct API and cache-inclusive costs, migration from GPT-6 Sol, Astra tradeoffs, and a reproducible evaluation plan
- [GPT-6.1 Sol vs Claude Opus 5.5](https://omniakey.com/blog/gpt-6-1-sol-vs-claude-opus-5-5): coding comparison with official benchmark context, direct API prices, reasoning and tool compatibility, cache-aware cost examples, and a reproducible evaluation plan
- [GPT-6 Sol review](https://omniakey.com/blog/gpt-6-sol): released model specifications, coding use cases, official API prices, long-context costs, and evaluation criteria
- [GPT-6 Luna review](https://omniakey.com/blog/gpt-6-luna-review): low-cost API review, batch economics, extraction and classification use cases, and output validation
- [GPT-6 Sol vs Luna](https://omniakey.com/blog/gpt-6-sol-vs-luna): model comparison, 20x token-price gap, equal-workload costs, and conditional routing economics
- [GPT-6 Sol vs Astra](https://omniakey.com/blog/gpt-6-sol-vs-astra): independent same-setting benchmark results, direct and gateway cost calculations, API tool boundaries, and workload selection
- [Nano Banana 2.1 vs Pro](https://omniakey.com/blog/nano-banana-2-1-vs-pro): current model IDs, Google Standard API costs at 1K/2K/4K, documented text and editing capabilities, and cost per accepted image; a sourced comparison, not a new generation benchmark
- [Nano Banana 2 vs Pro review](https://omniakey.com/blog/nano-banana-2-vs-pro-review): Google model IDs, image pricing, OmniaKey API steps, and matched generation and edit tests
- [GPT-Image 2.5 vs GPT-Image 2](https://omniakey.com/blog/gpt-image-2-5-vs-gpt-image-2): official model IDs, quality levels, output sizes, editing limits, and per-call pricing
- [GPT Image 2 vs Nano Banana 2](https://omniakey.com/blog/gpt-image-2-vs-nano-banana-2): cross-provider model IDs, image input and output controls, API parameter differences, gateway and provider pricing, and a fair acceptance-test protocol
- [DeepSeek V4.1 Flash review](https://omniakey.com/blog/deepseek-v4-1-flash-review): Official API prices, architecture, agent benchmarks, model IDs, and the V4 Pro migration
- [GPT-6 Astra setup](https://omniakey.com/blog/gpt-6-astra-setup): Responses API with Python, Node.js, and Codex
- [GPT-6 Astra review](https://omniakey.com/blog/gpt-6-astra-review): pricing, context window, and tool support
- [GPT-5.6 review](https://omniakey.com/blog/gpt-5-6-review): pricing and model comparison
- [Grok 4.6 API pricing](https://omniakey.com/blog/grok-4-6-api-pricing): rates and where Grok fits
- [Claude Code vs Cowork](https://omniakey.com/blog/claude-code-vs-cowork): repository delivery versus knowledge-work artifacts, access plans, permissions, and API-key boundaries
- [DeepSeek V4 Pro API pricing](https://omniakey.com/blog/deepseek-v4-pro-api-pricing): rates and positioning
- [DeepSeek V4 Flash Vision Exp](https://omniakey.com/blog/deepseek-v4-flash-vision-exp): image input and OCR
- [Transparent AI gateway billing](https://omniakey.com/blog/transparent-billing): reconcile per-request model, token, cache, latency, and cost records
- [WorkBuddy custom model setup](https://omniakey.com/blog/workbuddy-custom-model-setup): connect a third-party LLM API, add an exact model ID, and troubleshoot the connection
- [Claude Haiku 4.5 vs GPT-5.6 Luna](https://omniakey.com/blog/claude-haiku-4-5-vs-gpt-5-6-luna): small-model comparison
- [Claude API vs OpenAI API pricing](https://omniakey.com/blog/claude-api-pricing-vs-openai-api-pricing): direct token rates, cache writes, Batch, long-context multipliers, and completed-task cost
- [Claude Sonnet 5 vs GPT-5.6 Terra](https://omniakey.com/blog/claude-sonnet-5-vs-gpt-5-6-terra): mid-tier comparison
- [GLM-5.3 vs GLM-5.2 for coding](https://omniakey.com/blog/glm-5-3-vs-glm-5-2-for-coding): coding benchmarks, equal official rates, API migration and upgrade decision
- [DeepSeek Harness OmniaKey setup](https://omniakey.com/blog/deepseek-harness-omniakey-setup): harness configuration
- [Claude Fable 5 vs GPT-5.6 Sol](https://omniakey.com/blog/claude-fable-5-vs-gpt-5-6-sol): frontier-model comparison
- [Claude Fable 5.1 review](https://omniakey.com/blog/claude-fable-5-1): release changes, pricing, model IDs, migration errors, and Claude Code compatibility
- [Claude Fable 5.5 release status](https://omniakey.com/blog/claude-fable-5-5-release-date): dated official-source check, unconfirmed release date and API details, current Claude price baselines, and how to verify model access; no Fable 5.5 route or launch is asserted
- [Claude Fable 5 pricing](https://omniakey.com/blog/claude-fable-5-pricing): token rates
- [Cline custom API key setup](https://omniakey.com/blog/cline-custom-api-key): step-by-step configuration
- [Claude Code pricing](https://omniakey.com/blog/claude-code-pricing): subscription vs API cost
- [GLM-5.2 API pricing](https://omniakey.com/blog/glm-5-2-api-pricing): token rates
- [Reduce Claude Code token usage](https://omniakey.com/blog/reduce-claude-code-token-usage): caching and context tactics
- [Claude Code API without a subscription](https://omniakey.com/blog/claude-code-api-without-subscription): legitimate local pay-as-you-go API setup, billing paths, costs, and troubleshooting
- [Best GPT-5.6 model for Codex](https://omniakey.com/blog/best-gpt-5-6-model-for-codex): model selection guide
- [Claude Opus 5 vs GPT-5.6 Sol](https://omniakey.com/blog/claude-opus-5-vs-gpt-5-6-sol): frontier-model comparison
- [Claude Opus 5 vs Sonnet 5](https://omniakey.com/blog/claude-opus-5-vs-sonnet-5): capability and cost trade-off
- [Cursor usage limit](https://omniakey.com/blog/cursor-usage-limit): paid-plan BYOK workaround
- [Best model for Claude Code](https://omniakey.com/blog/best-model-for-claude-code): model selection guide
- [Claude Code vs Codex](https://omniakey.com/blog/claude-code-vs-codex): agent comparison
- [Seedance 2.5 review](https://omniakey.com/blog/seedance-2-5-review): evidence-based review
- [Claude Opus 5 review](https://omniakey.com/blog/claude-opus-5-review): capabilities and pricing
- [GPT-5.6 Codex compatibility kit](https://omniakey.com/blog/gpt-5-6-codex-compatibility-kit): configuration reference
- [OpenRouter alternative](https://omniakey.com/blog/openrouter-alternative): routing and pricing comparison

- [Nano Banana 2 Lite vs Nano Banana 2](https://omniakey.com/blog/nano-banana-2-lite-vs-nano-banana-2): API prices, 1K versus 4K output, editing constraints, and calculated cost per usable image; distinguishes Google direct billing from OmniaKey per-call pricing.
- [Nano Banana 2.1 review](https://omniakey.com/blog/nano-banana-2-1-review): documented typography, panoramic output and editing improvements; Google API cost calculations, Thinking defaults, and migration from Nano Banana 2. It also shows how to create supported Nano Banana 2 image tasks directly on OmniaKey with the asynchronous Media API, while keeping the exact 2.1 catalog status explicit.

- [GPT-6 Intelligent UI guide](https://omniakey.com/blog/gpt-6-intelligent-ui-guide): ChatGPT access, supported thinking levels, three original prompts with independently checked examples, usage limits, and the boundary between ChatGPT UI and model APIs.

## Optional

- [Chinese docs](https://docs.omniakey.com/cn/introduction): 中文文档首页
- [Chinese quick start](https://docs.omniakey.com/cn/quick-start): 中文快速上手
- [Chinese API reference](https://docs.omniakey.com/cn/api): 中文 API 参考
- [Chinese API overview](https://docs.omniakey.com/cn/api/overview): 中文 API 总览
- [Chinese Claude Code guide](https://docs.omniakey.com/cn/guides/claude-code-api-key): 中文 Claude Code 配置
- [Chinese Cursor guide](https://docs.omniakey.com/cn/guides/cursor-api-key): 中文 Cursor 配置
- [Russian docs](https://docs.omniakey.com/ru/introduction): документация на русском
- [Russian quick start](https://docs.omniakey.com/ru/quick-start): быстрый старт
- [Russian API reference](https://docs.omniakey.com/ru/api): справочник API
- [Russian API overview](https://docs.omniakey.com/ru/api/overview): обзор API
- [Russian Claude Code guide](https://docs.omniakey.com/ru/guides/claude-code-api-key): настройка Claude Code
- [Russian Cursor guide](https://docs.omniakey.com/ru/guides/cursor-api-key): настройка Cursor
- [Turkish docs](https://docs.omniakey.com/tr/introduction): Türkçe dokümantasyon
- [Turkish quick start](https://docs.omniakey.com/tr/quick-start): hızlı başlangıç
- [Turkish API reference](https://docs.omniakey.com/tr/api): API referansı
- [Turkish API overview](https://docs.omniakey.com/tr/api/overview): API genel bakış
- [Turkish Claude Code guide](https://docs.omniakey.com/tr/guides/claude-code-api-key): Claude Code kurulumu
- [Turkish Cursor guide](https://docs.omniakey.com/tr/guides/cursor-api-key): Cursor kurulumu
- [Spanish docs](https://docs.omniakey.com/es/introduction): documentación en español
- [Spanish quick start](https://docs.omniakey.com/es/quick-start): inicio rápido
- [Spanish API reference](https://docs.omniakey.com/es/api): referencia de la API
- [Spanish API overview](https://docs.omniakey.com/es/api/overview): visión general de la API
- [Spanish Claude Code guide](https://docs.omniakey.com/es/guides/claude-code-api-key): configuración de Claude Code
- [Spanish Cursor guide](https://docs.omniakey.com/es/guides/cursor-api-key): configuración de Cursor
- [Terms](https://omniakey.com/terms): terms of service
- [Privacy](https://omniakey.com/privacy): privacy policy
- [Refund](https://omniakey.com/refund): refund policy

## September 28, 2026 guides

- [Jev API Guide](https://omniakey.com/blog/jev-api-guide): Native decision requests, typed responses, limits and HTTP client examples.
- [DeepSeek API Pricing](https://omniakey.com/blog/deepseek-api-pricing): Direct Flash and Pro token prices, cache accounting, UTC peak windows and gateway billing boundaries.
- [Claude Opus 5.5 vs Sonnet 5](https://omniakey.com/blog/claude-opus-5-5-vs-sonnet-5): Coding decisions, standard and cached costs, effort and latency evidence.

## Developer tools

- [MCP](https://omniakey.com/mcp): Local stdio tools for coding agents.
- [Skill](https://omniakey.com/skill): Install the omniakey-models workflow.
- [CLI](https://omniakey.com/cli): Terminal and CI workflows with JSON output.
- [Source](https://github.com/omniakey-com/omniakey-devtools): CLI, MCP, Skill and installation documentation.