# Tokely

> A hosted AI API for supported text, image, video and audio models, with one Tokely key and USD balance. API base URL: https://api.tokely.me/v1.

## Product facts and boundaries

OpenAI-compatible Chat Completions and a stateless Responses subset for supported models. Compatible models accept inline PNG/JPEG/WebP images and PDFs. Text API JSON bodies are limited to 16 MB, including base64; output is capped at the model limit or 262,144 tokens. Remote image/file URLs, uploaded file IDs and audio/video input are not supported. Supported media uses image generation, speech or asynchronous media tasks, depending on the model. Check the exact model page for parameters and rates.
Account routing modes: Balanced, Lowest cost, Fastest, Most reliable and Fixed. Compatible route fallback is bounded by accepted-task and partial-response safeguards; redundancy is not available for every model.
Account budgets, key restrictions, usage/logs and recorded savings where comparable request-time baselines exist. Hosted crypto top-ups. BYOK, team RBAC, SSO, Batch API and speech transcription are not currently offered. OpenAI embedding models are served at /v1/embeddings. Claude models also support the native Anthropic Messages endpoint (/v1/messages) for Claude Code and the Anthropic SDK, with prompt caching.
Tokely operates the price index. It is a first-party source, not an independent ranking. Prices have separate source dates; no market-wide lowest-cost, measured quality or uptime claim is made.

## Pricing and comparisons

- [AI API resources](https://tokely.me/resources): Index of every pricing, model, gateway, comparison and provider page.
- [AI API Pricing — Model Rates & Discounts](https://tokely.me/pricing): Explore Tokely prices and discounts for text, images, video and audio. One API key and USD balance. Pay as you go. No subscription.
- [AI API Cost Calculator — Estimate Token Spend](https://tokely.me/calculator): Calculate your AI API costs with Tokely model prices. Adjust input, output and cached tokens, then compare your estimate with model developer prices.
- [Cheap AI API — Text, Image & Video Pricing](https://tokely.me/cheapest-ai-api): Find lower-cost text, image and video APIs on Tokely. Compare model prices and discounts, then connect through one OpenAI-compatible API.
- [Cheap LLM API — Token Prices & Discounts](https://tokely.me/cheapest-llm-api): Compare LLM input, output and cached-token prices on Tokely. Explore GPT, Claude and Gemini with one OpenAI-compatible API. Pay as you go.
- [Cheap Claude API — Sonnet & Opus Pricing](https://tokely.me/cheap-claude-api): Access Claude Sonnet and Opus through Tokely. Compare input, output and cached-token prices with Anthropic rates. Pay as you go. No subscription.
- [Cheap GPT API — Model Prices & Discounts](https://tokely.me/cheap-gpt-api): Access GPT models through Tokely with your OpenAI SDK. Compare input, output and cached-token prices with OpenAI rates. Pay as you go.
- [Cheap Gemini API — Model Prices & Discounts](https://tokely.me/cheap-gemini-api): Access Gemini text models through Tokely. Compare input, output and cached-token prices with Google rates. One API key. No subscription.
- [Cheap AI Image API — Prices per Image](https://tokely.me/cheap-ai-image-api): Generate images with GPT Image, FLUX and Gemini through Tokely. Compare prices per image and discounts for each model. Pay as you go.
- [Cheap AI Video API — Model Prices & Discounts](https://tokely.me/cheap-ai-video-api): Generate videos with Grok Imagine, Seedance and Veo through Tokely. Compare prices per second, clip costs and model discounts. No subscription.
- [AI Audio API Pricing — Speech & Music](https://tokely.me/cheap-ai-audio-api): Explore speech and music API pricing on Tokely. Check model availability, billing units and generation settings before you connect.
- [Claude Code API: run Claude Code on a cheaper key](https://tokely.me/claude-code-api): Use Claude Code with a Tokely API key. Native Anthropic Messages endpoint, streaming, tools and prompt caching, billed per token from one balance with no subscription.
- [LLM gateway: one API for every model](https://tokely.me/llm-gateway): A hosted LLM gateway with one OpenAI-compatible endpoint, account routing modes, bounded fallback, spending limits and request logs. Text, image, video and audio from one balance.
- [OpenAI-compatible API for Claude, Gemini, GPT and more](https://tokely.me/openai-compatible-api): Keep the OpenAI SDK and call Claude, Gemini, GPT, Kimi, GLM and other supported models. Change the base URL, key and model ID.
- [Claude API pricing by model](https://tokely.me/claude-api-pricing): Claude Opus, Sonnet, Haiku and Fable API prices per million input, output and cached-input tokens, compared with Anthropic's published rates and Tokely's rates.
- [A cost-focused OpenRouter alternative](https://tokely.me/openrouter-alternative): Compare model rates, platform fees, supported endpoints and migration effort before moving from OpenRouter to Tokely. One balance for text, image, video and audio.
- [OpenRouter pricing explained](https://tokely.me/openrouter-pricing): How OpenRouter charges: pass-through model prices, a 5.5% or 8% credit purchase fee, BYOK fees and free-model limits. Compare the total with a gateway that has no platform fee.
- [OpenRouter alternatives: choose by the job](https://tokely.me/openrouter-alternatives): A practical comparison of hosted gateways, direct model hosts and a self-hosted proxy. Compare Tokely, Requesty, Together AI, Fireworks AI, DeepInfra and LiteLLM by your requirements.
- [LLM routing and fallback](https://tokely.me/llm-routing): Route each LLM request by cost, speed or reliability, with bounded fallback to another supplier route for the same model. Set a mode per account, key or request.
- [LLM observability: requests, usage and cost](https://tokely.me/llm-observability): Request IDs, attempt-level logs, token and cost usage, exports, spending limits and signed webhooks for every LLM call through one gateway.
- [Reduce LLM costs in production](https://tokely.me/llm-cost-optimization): Practical ways to cut LLM API spend: per-model discounts, prompt caching billed correctly, lowest-cost routing, right-sized models and hard budgets per key.
- [An AI API for SaaS products](https://tokely.me/ai-api-for-saas): Ship AI features on one OpenAI-compatible key: per-key budgets and model allow-lists, routing per key, usage per customer feature, signed webhooks and one prepaid balance.
- [OpenAI API alternative with lower rates](https://tokely.me/openai-api-alternative): Keep the OpenAI SDK and call GPT, Claude, Gemini, Kimi, DeepSeek and more at published per-model rates. Change the base URL and key; no platform fee.
- [DeepSeek API: V4 Flash and V4 Pro pricing](https://tokely.me/deepseek-api): Call DeepSeek V4.1 Flash and V4 Pro through an OpenAI-compatible API. Per-token prices, cache-read rates, tools, JSON and streaming from one Tokely key.
- [Grok API pricing: Grok 4.3 to 4.7](https://tokely.me/grok-api): Use xAI Grok models through an OpenAI-compatible API. Compare Grok 4.3, 4.5, 4.6 and 4.7 per-token prices and call them with one Tokely key.
- [Kimi API: Kimi K3 and K2.7 Code](https://tokely.me/kimi-api): Call Moonshot Kimi K3 and Kimi K2.7 Code through an OpenAI-compatible API, with tools, JSON, streaming and prompt caching from one key.
- [Gemini vs Claude: API price and fit](https://tokely.me/gemini-vs-claude): Compare Gemini and Claude API models side by side: per-token prices, cache rates, input types and integration. Pick per task, then test both on one key.
- [Claude vs GPT: API price and fit](https://tokely.me/claude-vs-gpt): Claude and GPT API models compared by per-token price, cache rate, endpoints and tooling. Choose per task and switch between them with one OpenAI-compatible key.
- [Best LLM for coding: how to choose](https://tokely.me/best-llm-for-coding): Coding models compared by price and fit: Claude Sonnet and Opus, GPT Codex, Kimi K2.7 Code, DeepSeek V4 Pro and Gemini Pro. Test them on one key in your own tools.
- [Use Cline, Roo, Kilo, OpenCode and Continue with one API key](https://tokely.me/coding-agents-api): Connect AI coding agents to Claude, GPT, Kimi K2.7 Code, DeepSeek and Gemini through Tokely's OpenAI-compatible endpoint. Copy-paste settings for each tool.
- [DeepSeek, Kimi, Qwen and MiniMax API](https://tokely.me/chinese-ai-models-api): Call DeepSeek V4, Kimi K3 and K2.7 Code, Qwen 3.8 Flash and MiniMax M3 through one OpenAI-compatible API key. Compare per-token prices in one table.
- [Embeddings API at half of OpenAI's price](https://tokely.me/embeddings-api): OpenAI text-embedding-3-small, text-embedding-3-large and ada-002 through an OpenAI-compatible /v1/embeddings endpoint, at 50% of OpenAI's list price.
- [Nano Banana API: Gemini image generation pricing](https://tokely.me/nano-banana-api): Generate images with Google's Nano Banana family (Gemini Flash Image and Gemini 3 Pro Image) through Tokely's API. Per-image prices, one key and balance.
- [Seedance API: ByteDance video generation pricing](https://tokely.me/seedance-api): Generate video with ByteDance Seedance 1.0 Pro and 1.5 Pro through Tokely's asynchronous media API. Per-task prices and one key for text, image and video.
- [Seedream API: ByteDance image generation pricing](https://tokely.me/seedream-api): Generate images with ByteDance Seedream 3.0 and Seedream 5.0 Pro through Tokely's API, priced per image, on the same key as text and video models.
- [GPT Image API pricing](https://tokely.me/gpt-image-api): OpenAI GPT Image 1.5, GPT Image 2 and GPT Image 2.5 variants through Tokely's API. Per-image prices and one key for text, image and video.
- [Veo API pricing: Google Veo 3.1 and 3.1 Fast](https://tokely.me/veo-api): Generate video with Google Veo 3.1 and Veo 3.1 Fast through Tokely's asynchronous media API. Per-task prices on one key and balance.
- [Use OpenClaw with Tokely: one key for every model](https://tokely.me/openclaw-api): Connect OpenClaw to Claude, GPT, Kimi, DeepSeek and Gemini through Tokely. Add a custom provider with the OpenAI-compatible or Anthropic-compatible endpoint.
- [AI API Price Index — Download Model Rates](https://tokely.me/ai-api-pricing): Download Tokely AI model prices in JSON or CSV. Explore token rates, image and video costs, model developer comparisons and dated sources.
- [How We Compare AI API Prices & Discounts](https://tokely.me/pricing-methodology): See how Tokely calculates AI API price discounts. Review model matching, comparison formulas, source dates, billing units and exclusions.
- [Gateway comparisons](https://tokely.me/compare): Product, pricing model and migration tradeoffs.
- [OpenRouter vs Tokely](https://tokely.me/compare/openrouter-vs-tokely): Catalog breadth or lower per-model price?
- [Requesty vs Tokely](https://tokely.me/compare/requesty-vs-tokely): Unit cost or platform governance?
- [Together AI vs Tokely](https://tokely.me/compare/together-ai-vs-tokely): Gateway or your own deployment?
- [Fireworks AI vs Tokely](https://tokely.me/compare/fireworks-ai-vs-tokely): Serving control or multi-provider access?
- [DeepInfra vs Tokely](https://tokely.me/compare/deepinfra-vs-tokely): Direct host or a managed gateway?
- [Portkey vs Tokely](https://tokely.me/compare/portkey-vs-tokely): Control plane or bought inference?
- [Helicone vs Tokely](https://tokely.me/compare/helicone-vs-tokely): Observability tool or API provider?
- [Cloudflare AI Gateway vs Tokely](https://tokely.me/compare/cloudflare-ai-gateway-vs-tokely): Edge gateway or lower model rates?
- [LiteLLM vs Tokely](https://tokely.me/compare/litellm-vs-tokely): Operate the proxy or buy API access?

## Sources and downloadable data

- [JSON price index](https://tokely.me/data/ai-api-prices.json): Published token rates, exact-model reference URLs, dates, media units and scope.
- [Text price CSV](https://tokely.me/data/ai-api-prices.csv): Same text rates in CSV.
- [October 2026 snapshot](https://tokely.me/insights/ai-api-pricing-october-2026): Reproducible price comparison; no historical trend claim.

## Field guide

- [Tokely blog](https://tokely.me/blog): Practical AI API guides, pricing and production engineering.
- [LLM Gateway vs Direct API: When Do You Need One?](https://tokely.me/blog/llm-gateway-vs-direct-api): A direct API is usually enough when your application uses one provider and your team can manage its credentials, limits and billing.
- [How to Add Model Fallback to Production LLM Apps](https://tokely.me/blog/llm-model-fallback): LLM fallback sends a failed request to an approved alternative when the original route cannot complete it.
- [LLM Routing Strategies: Balancing Cost, Latency and Quality](https://tokely.me/blog/llm-routing-strategies): LLM routing chooses which model or provider should handle a request.
- [How to Reduce LLM API Costs Without Sacrificing Quality](https://tokely.me/blog/reduce-llm-api-costs): Reduce LLM API costs by removing unnecessary calls, matching model capability to the task, reusing eligible prompt prefixes and controlling output length.
- [Semantic Caching vs Prompt Caching: How They Work and When They Save Money](https://tokely.me/blog/semantic-caching-vs-prompt-caching): Prompt caching reuses the processing of repeated input while generating a new response.
- [Managing Rate Limits Across Multiple LLM Providers](https://tokely.me/blog/llm-rate-limits): Manage LLM rate limits with queues, token-aware admission control, bounded retries and separate budgets for each provider or model.
- [LLM Observability: What to Track Beyond Tokens and Cost](https://tokely.me/blog/llm-observability): LLM observability should connect each user action to its model calls, token usage, cost, latency, errors and accepted outcome.
- [BYOK vs Managed Keys for AI SaaS: Who Owns the API Bill?](https://tokely.me/blog/byok-vs-managed-keys): BYOK, or bring your own key, means a customer supplies credentials for their provider account.
- [Self-Hosted LiteLLM vs a Managed Gateway: Comparing Total Cost](https://tokely.me/blog/litellm-vs-managed-gateway): Self-hosted LiteLLM makes sense when your team needs control over gateway configuration and can operate the service reliably.
- [Multi-Model Architecture for AI SaaS: A Practical Design](https://tokely.me/blog/multi-model-ai-saas-architecture): A multi-model SaaS architecture separates product workflows from model-specific API calls.
- [Per-Customer Usage Tracking and Billing for AI Features](https://tokely.me/blog/ai-usage-tracking-billing): Track AI usage at the customer and product-action level, then reconcile it with provider records.
- [Choosing an OpenAI-Compatible Provider for Production](https://tokely.me/blog/openai-compatible-provider-production): Choose an OpenAI-compatible provider by checking the exact endpoints and features your application uses, then evaluating billing, reliability and data handling.
- [What Is an AI API Aggregator?](https://tokely.me/blog/what-is-ai-api-aggregator): An AI API aggregator gives developers access to models from multiple providers through a shared service.
- [OpenRouter Alternatives in 2026: AI Aggregators and Gateways Compared](https://tokely.me/blog/openrouter-alternatives): The right OpenRouter alternative depends on what you want to replace: consolidated model access, provider routing, operational controls or the entire intermediary layer.
- [Claude Code With a Cheaper API Provider: Connecting to Tokely](https://tokely.me/blog/claude-code-cheaper-api): Claude Code can use Tokely through its Anthropic-compatible Messages endpoint.
- [DeepSeek V4, GLM 5.3 and Kimi K3: Finding the Cheapest API Route](https://tokely.me/blog/deepseek-glm-kimi-api-pricing): The cheapest route to DeepSeek V4, GLM 5.3 or Kimi K3 depends on the exact model, cached-input share and service conditions.
- [Free AI Models via API: What the Limits Actually Mean](https://tokely.me/blog/free-ai-models-api): Free AI API access usually means a limited model selection, a usage allowance or promotional credits.
- [AI API Pricing Explained: Input, Output, Caching and Hidden Costs](https://tokely.me/blog/ai-api-pricing-explained): Most text AI APIs charge separately for input and output tokens.
- [AI API Discounts and Startup Credits: How to Compare the Real Savings](https://tokely.me/blog/ai-api-startup-credits): Startup credits reduce cash spending for a limited period; discounted API rates reduce the price of eligible usage.
- [Is a Cheap AI API Reliable Enough for Production?](https://tokely.me/blog/cheap-ai-api-reliability): A cheap AI API can be suitable for production, but price alone does not establish reliability.
- [Security and Data Privacy With Third-Party AI APIs](https://tokely.me/blog/third-party-ai-api-security): A third-party AI API adds parties to the path between your application and a model.
- [Cheapest AI APIs in 2026: Compare Prices per Million Tokens](https://tokely.me/blog/cheapest-ai-apis): The cheapest AI API depends on the model and your input-to-output ratio.
- [OpenRouter vs Tokely: API Pricing and Features Compared](https://tokely.me/blog/openrouter-vs-tokely): OpenRouter and Tokely both offer access to multiple AI models, but the useful comparison is your required feature set and total workload cost.
- [Claude API Pricing Explained: Input, Output and Caching](https://tokely.me/blog/claude-api-pricing): Claude API cost depends on the selected model and separate charges for input, output and caching.
- [How to Use the OpenAI SDK With a Compatible API](https://tokely.me/blog/openai-sdk-compatible-api): To use the OpenAI SDK with Tokely, configure three values: the Tokely base URL, a Tokely API key and a model ID supported by the endpoint you call.
- [Gemini API vs Claude API: Comparing Price, Speed and Quality](https://tokely.me/blog/gemini-vs-claude-api): Compare Gemini and Claude at the model and task level, not as two undifferentiated brands.
- [AI API Cost Calculator: How to Estimate Monthly Spend](https://tokely.me/blog/ai-api-cost-calculator): Estimate monthly AI API spending by multiplying user actions by model calls per action, then pricing the input and output of each call.
- [Best AI Models for Coding: How to Compare Price and Performance](https://tokely.me/blog/best-coding-models-price-performance): The best coding model is the one that completes your repository tasks at an acceptable total cost and review burden.
- [What Are Embedding Models? API Pricing and Practical Uses](https://tokely.me/blog/embedding-models-pricing): An embedding model converts text into a numerical vector that represents useful relationships between inputs.
- [What Is Tokely? One API for GPT, Claude, Gemini and 130+ Models](https://tokely.me/blog/what-is-tokely): Tokely (tokely.me) is a hosted LLM gateway: one OpenAI-compatible API key and one prepaid balance for 137 AI models from 14 developers, at rates up to 85% below official prices.
- [Are AI APIs Getting Cheaper? Official Prices of 66 Models, 2024-2026](https://tokely.me/blog/ai-api-price-trends-2026): Study of 66 AI models at current official prices: the newest flagship lists output at 6.25x a 2025 flagship, the $10 tier holds, and cached input is 10 to 50x cheaper.
- [Claude API 2026-10-21: What Breaks and How to Upgrade](https://tokely.me/blog/claude-api-2026-10-21-migration): Anthropic's API version 2026-10-21 removes temperature, top_p and top_k, changes SDK auth to Bearer and renames Admin API fields. What breaks and how to upgrade.
- [DeepSeek Banned From U.S. Federal Agencies: Who It Affects and What to Use Instead](https://tokely.me/blog/deepseek-federal-ban-alternatives): What the DHS order on DeepSeek covers, who it affects, and which models cost less than DeepSeek V4.1 Flash at official list prices.
- [Gemini 4 Argon Pricing vs Claude Sonnet 5.5 and GPT-6.1 Sol: Same Price Until December 1](https://tokely.me/blog/gemini-4-argon-pricing): Gemini 4 Argon lists at $2 input and $10 output per 1M tokens, the same as Claude Sonnet 5.5 and GPT-6.1 Sol, until Google doubles it on December 1, 2026.
- [How to Review AI-Written Code: Lessons From the Claude Code Team and Hacker News](https://tokely.me/blog/how-to-review-ai-written-code): A practical review process for code written by coding agents, from the Claude Code team's seven review layers and risk tiers to what 748 Hacker News comments recommend.
- [Kimi K3.1 Pro API Pricing: Half of Claude Opus 5.5, Same as Sonnet 5.5](https://tokely.me/blog/kimi-k3-1-pro-pricing): Kimi K3.1 Pro lists at $2 input and $10 output per 1M tokens: a third below Kimi K3, half of Claude Opus 5.5, level with Sonnet 5.5, GPT-6.1 Sol and Argon.
- [Agent Skills Moves to the Agentic AI Foundation: What Developers Need to Know](https://tokely.me/blog/agent-skills-agentic-ai-foundation): On October 22, 2026, Agent Skills moved to the Linux Foundation's Agentic AI Foundation. What changes, how SKILL.md works and how to run skills on any model.
- [DeepSeek Open-Sourced the Engine Behind Its API. Is Self-Hosting V4.1 Flash Cheaper Than Paying per Token?](https://tokely.me/blog/deepseek-deepinfer-self-hosting-cost): DeepSeek open-sourced DeepInfer. The smallest efficient V4.1 Flash cluster rents for $880K a month and beats the $0.60 API price only above 76% utilization.
- [Web Search API Pricing for AI Agents: 8 Providers Compared (2026)](https://tokely.me/blog/web-search-api-pricing-2026): Study of 13 web search API tiers from 8 providers: $0.75 to $35 per 1,000 searches, a 47x spread, and model-native search tools at 2 to 2.8x the median standalone API.
- [Are Chinese AI Models Getting More Expensive? API Prices by Generation (2026)](https://tokely.me/blog/chinese-ai-model-api-prices-2026): Official API prices for 24 models from DeepSeek, Qwen, Kimi, GLM and MiniMax: DeepSeek peak output at 4.7x its August 1 rate, a 2.15x to 2.34x generation premium for GLM and Kimi, and Kimi K3.1 Pro at the GPT-6.1 Sol price.
- [Claude Account Bans After ID Verification: What Happened in October 2026](https://tokely.me/blog/claude-account-bans-identity-verification-2026): Users with Russian passports report Claude bans right after Persona identity verification, including Pro and Max subscribers living in supported countries. What is known, the earlier waves, refunds and appeals.

## Documentation

- [One API. Your next idea.](https://tokely.me/docs): Connect your app to AI models through one familiar API. This is your guide to building with Tokely.
- [From key to first response.](https://tokely.me/docs/quickstart): Make your first Tokely request in three steps.
- [API keys & authentication](https://tokely.me/docs/authentication): Connect securely with a key from your Tokely dashboard.
- [Find the right model.](https://tokely.me/docs/models): Choose an available model and use its exact ID in your request.
- [Try a model. Start a conversation.](https://tokely.me/docs/playground): Use the Tokely Playground to test prompts before writing code.
- [Bring your existing app.](https://tokely.me/docs/migration): Connect an existing OpenAI-compatible application to Tokely.
- [Chat Completions](https://tokely.me/docs/chat-completions): Generate responses from text messages and supported inline images or PDFs.
- [Responses](https://tokely.me/docs/responses): Use stateless text and supported multimodal input with the Responses API.
- [Embeddings](https://tokely.me/docs/embeddings): Turn text into vectors for search, RAG, clustering and classification with the OpenAI-compatible /v1/embeddings endpoint.
- [Claude Code and the Messages API](https://tokely.me/docs/claude-code): Point Claude Code or the Anthropic SDK at Tokely. Use the native /v1/messages endpoint with streaming, tools and prompt caching.
- [Images and PDFs](https://tokely.me/docs/multimodal): Send inline images and PDFs to compatible models through Chat Completions or Responses.
- [Images, video & audio](https://tokely.me/docs/media): Generate media with your Tokely API key and account balance.
- [Errors & limits](https://tokely.me/docs/errors): Diagnose failed requests and keep your integration within service limits.
- [Let the response flow.](https://tokely.me/docs/streaming): Display generated text as it arrives with server-sent events.
- [Connect models to your tools.](https://tokely.me/docs/tools): Let a model request a function, then execute it in your application.
- [Get JSON your app can use.](https://tokely.me/docs/structured-output): Request JSON output or a defined schema from a compatible model.
- [Understand your usage.](https://tokely.me/docs/billing): Model prices, account balance and output reservations.
- [Webhooks: signed events & retries](https://tokely.me/docs/webhooks): Tokely webhook events, HMAC signature verification, delivery retries and replay behavior.

## Catalog and model developers

- [text models](https://tokely.me/models/text): Enabled models, rates and contracts.
- [image models](https://tokely.me/models/image): Enabled models, rates and contracts.
- [video models](https://tokely.me/models/video): Enabled models, rates and contracts.
- [audio models](https://tokely.me/models/audio): Enabled models, rates and contracts.
- [OpenAI](https://tokely.me/providers/openai): Compare GPT model versions, input/output rates and Chat Completions or Responses support. Choose a model by the endpoint and limits your application needs.
- [Anthropic](https://tokely.me/providers/anthropic): Compare Claude Sonnet and Opus API costs. Use Claude through Tokely's OpenAI-compatible interface or the native Anthropic Messages endpoint used by Claude Code.
- [Google](https://tokely.me/providers/google): Compare Gemini text and image options. Image variants have their own fixed settings and billing units; token prices apply only to the listed text endpoints.
- [DeepSeek](https://tokely.me/providers/deepseek): Compare DeepSeek input, output and cached-input costs. The input/output mix can change which route is cheaper, so use the same workload on both sides.
- [Grok](https://tokely.me/providers/xai): Explore Grok text and Imagine media models. Text pricing uses tokens; video pricing applies to the specific duration and resolution on each model page.
- [Qwen](https://tokely.me/providers/qwen): Explore Qwen text and image models. Compare text API capabilities separately from fixed image-generation variants.
- [Moonshot](https://tokely.me/providers/moonshot): Compare Kimi text models and their published token rates. Check the exact version, context window and supported parameters before switching an application.
- [Doubao](https://tokely.me/providers/bytedance): Explore ByteDance Seedream image and Seedance video variants. Match reference images, duration and resolution before comparing generation prices.
- [Black Forest Labs](https://tokely.me/providers/black-forest-labs): Explore FLUX image-generation variants. Each published variant specifies image/reference requirements and a bounded per-generation charge.
- [Kling](https://tokely.me/providers/kling): Explore Kling video API variants with explicit duration, resolution and input requirements. A lower price for a different video length is a different product.
- [MiniMax](https://tokely.me/providers/minimax): Explore MiniMax text, video and audio models. Use the billing unit and parameter bounds on each model page, not one cross-modality price.
- [Suno](https://tokely.me/providers/suno): Explore Suno music-generation API variants. Check the published music duration, mode and output contract; music-generation costs cannot be compared directly with speech token prices.

## Enabled models

- [Claude Fable 5](https://tokely.me/models/claude-fable-5): ID claude-fable-5; Text; /v1/chat/completions; $7.5 input / $37.5 output per 1M tokens.
- [Claude Opus 4.6](https://tokely.me/models/claude-opus-4-6): ID claude-opus-4-6; Text; /v1/chat/completions; $3.75 input / $18.75 output per 1M tokens.
- [Claude Opus 4.7](https://tokely.me/models/claude-opus-4-7): ID claude-opus-4-7; Text; /v1/chat/completions; $3.75 input / $18.75 output per 1M tokens.
- [Claude Opus 4.8](https://tokely.me/models/claude-opus-4-8): ID claude-opus-4-8; Text; /v1/chat/completions; $3.75 input / $18.75 output per 1M tokens.
- [Claude Opus 5](https://tokely.me/models/claude-opus-5): ID claude-opus-5; Text; /v1/chat/completions; $3.75 input / $18.75 output per 1M tokens.
- [Claude Opus 5.5](https://tokely.me/models/claude-opus-5-5): ID claude-opus-5-5; Text; /v1/chat/completions; $3 input / $15 output per 1M tokens.
- [Claude Sonnet 4.6](https://tokely.me/models/claude-sonnet-4-6): ID claude-sonnet-4-6; Text; /v1/chat/completions; $2.25 input / $11.25 output per 1M tokens.
- [Claude Sonnet 5](https://tokely.me/models/claude-sonnet-5): ID claude-sonnet-5; Text; /v1/chat/completions; $1.5 input / $7.5 output per 1M tokens.
- [Claude Sonnet 5.5](https://tokely.me/models/claude-sonnet-5-5): ID claude-sonnet-5-5; Text; /v1/chat/completions; $1.5 input / $7.5 output per 1M tokens.
- [DeepSeek V4.1 Flash](https://tokely.me/models/deepseek-v4.1-flash): ID deepseek-v4.1-flash; Text; /v1/chat/completions; $0.2 input / $0.8 output per 1M tokens.
- [Seedance 1.0 Pro](https://tokely.me/models/doubao-seedance-1-0-pro-250528): ID doubao-seedance-1-0-pro-250528; Video; /v1/media/tasks; $0.02187 / second.
- [Seedance 1.5 Pro](https://tokely.me/models/doubao-seedance-1-5-pro-251215): ID doubao-seedance-1-5-pro-251215; Video; /v1/media/tasks; $0.01116 / second.
- [Seedream 3.0](https://tokely.me/models/doubao-seedream-3-0-t2i-250415): ID doubao-seedream-3-0-t2i-250415; Image; /v1/images/generations; $0.027 / image.
- [Seedream 5.0 Pro](https://tokely.me/models/doubao-seedream-5-0-pro-260628): ID doubao-seedream-5-0-pro-260628; Image; /v1/images/generations; $0.0525 / image.
- [Gemini 2.5 Flash](https://tokely.me/models/gemini-2.5-flash): ID gemini-2.5-flash; Text; /v1/chat/completions; $0.21 input / $1.75 output per 1M tokens.
- [Nano Banana](https://tokely.me/models/gemini-2.5-flash-image): ID gemini-2.5-flash-image; Image; /v1/images/generations; $0.0312 / image.
- [Nano Banana Pro](https://tokely.me/models/gemini-3-pro-image-preview): ID gemini-3-pro-image-preview; Image; /v1/images/generations; $0.0938 / image.
- [Nano Banana 2](https://tokely.me/models/gemini-3.1-flash-image-preview): ID gemini-3.1-flash-image-preview; Image; /v1/images/generations; $0.0469 / image.
- [Gemini 3.1 Flash Lite](https://tokely.me/models/gemini-3.1-flash-lite): ID gemini-3.1-flash-lite; Text; /v1/chat/completions; $0.175 input / $1.05 output per 1M tokens.
- [Gemini 3.5 Flash](https://tokely.me/models/gemini-3.5-flash): ID gemini-3.5-flash; Text; /v1/chat/completions; $1.05 input / $6.3 output per 1M tokens.
- [Gemini 3.5 Flash Lite](https://tokely.me/models/gemini-3.5-flash-lite): ID gemini-3.5-flash-lite; Text; /v1/chat/completions; $0.21 input / $1.75 output per 1M tokens.
- [Gemini 3.7 Flash](https://tokely.me/models/gemini-3.7-flash): ID gemini-3.7-flash; Text; /v1/chat/completions; $0.525 input / $2.625 output per 1M tokens.
- [Gemini 3.8 Flash](https://tokely.me/models/gemini-3.8-flash): ID gemini-3.8-flash; Text; /v1/chat/completions; $0.525 input / $2.625 output per 1M tokens.
- [GPT-5.4](https://tokely.me/models/gpt-5.4): ID gpt-5.4; Text; /v1/responses; $1.75 input / $10.5 output per 1M tokens.
- [GPT-5.5](https://tokely.me/models/gpt-5.5): ID gpt-5.5; Text; /v1/chat/completions; $2 input / $12 output per 1M tokens.
- [GPT-5.6 Luna](https://tokely.me/models/gpt-5.6-luna): ID gpt-5.6-luna; Text; /v1/chat/completions; $0.14 input / $0.84 output per 1M tokens.
- [GPT-5.6 Sol](https://tokely.me/models/gpt-5.6-sol): ID gpt-5.6-sol; Text; /v1/chat/completions; $2.8 input / $14 output per 1M tokens.
- [GPT-5.6 Terra](https://tokely.me/models/gpt-5.6-terra): ID gpt-5.6-terra; Text; /v1/chat/completions; $1.4 input / $8.4 output per 1M tokens.
- [GPT-6 Astra](https://tokely.me/models/gpt-6-astra): ID gpt-6-astra; Text; /v1/chat/completions; $4 input / $20 output per 1M tokens.
- [GPT-6 Luna](https://tokely.me/models/gpt-6-luna): ID gpt-6-luna; Text; /v1/chat/completions; $0.07 input / $0.35 output per 1M tokens.
- [GPT-6 Sol](https://tokely.me/models/gpt-6-sol): ID gpt-6-sol; Text; /v1/chat/completions; $1.4 input / $7 output per 1M tokens.
- [GPT-6.1 Sol](https://tokely.me/models/gpt-6.1-sol): ID gpt-6.1-sol; Text; /v1/chat/completions; $0.8 input / $4 output per 1M tokens.
- [GPT Image 1.5](https://tokely.me/models/gpt-image-1.5): ID gpt-image-1.5; Image; /v1/images/generations; $0.0306 / image.
- [GPT Image 2](https://tokely.me/models/gpt-image-2): ID gpt-image-2; Image; /v1/images/generations; $0.0054 / image.
- [GPT Image 2.5 Flare](https://tokely.me/models/gpt-image-2.5-flare): ID gpt-image-2.5-flare; Image; /v1/images/generations; $0.046144 / image.
- [GPT Image 2.5 Sunburst](https://tokely.me/models/gpt-image-2.5-sunburst): ID gpt-image-2.5-sunburst; Image; /v1/images/generations; $0.046144 / image.
- [Grok 4.3](https://tokely.me/models/grok-4.3): ID grok-4.3; Text; /v1/chat/completions; $0.1875 input / $0.375 output per 1M tokens.
- [Grok 4.5](https://tokely.me/models/grok-4.5): ID grok-4.5; Text; /v1/chat/completions; $1.4 input / $4.2 output per 1M tokens.
- [Grok 4.6](https://tokely.me/models/grok-4.6): ID grok-4.6; Text; /v1/chat/completions; $1.4 input / $4.2 output per 1M tokens.
- [Grok 4.7](https://tokely.me/models/grok-4.7): ID grok-4.7; Text; /v1/chat/completions; $1.4 input / $4.2 output per 1M tokens.
- [Grok Imagine Video](https://tokely.me/models/grok-imagine-video): ID grok-imagine-video; Video; /v1/media/tasks; $0.018 / second.
- [Kimi K3](https://tokely.me/models/kimi-k3): ID kimi-k3; Text; /v1/chat/completions; $2.7 input / $13.5 output per 1M tokens.
- [Qwen Image 3.0](https://tokely.me/models/qwen-image-3.0): ID qwen-image-3.0; Image; /v1/images/generations; $0.036 / image.
- [Qwen Image 3.0 Pro](https://tokely.me/models/qwen-image-3.0-pro): ID qwen-image-3.0-pro; Image; /v1/images/generations; $0.048 / image.
- [Veo 3.1](https://tokely.me/models/veo_3_1): ID veo_3_1; Video; /v1/media/tasks; $0.16 / second.
- [Veo 3.1 Fast](https://tokely.me/models/veo_3_1-fast): ID veo_3_1-fast; Video; /v1/media/tasks; $0.07 / second.
- [Wan 2.5 Image to Video](https://tokely.me/models/wan2.5-i2v-preview): ID wan2.5-i2v-preview; Video; /v1/media/tasks; $0.09 / second.
- [Wan 2.6 Image to Video](https://tokely.me/models/wan2.6-i2v): ID wan2.6-i2v; Video; /v1/media/tasks; $0.09 / second.
- [FLUX.1 Kontext Pro](https://tokely.me/models/flux.1-kontext-pro): ID flux.1-kontext-pro; Image; /v1/images/generations; $0.036 / image.
- [HappyHorse 1.0](https://tokely.me/models/happyhorse-1.0-i2v): ID happyhorse-1.0-i2v; Video; /v1/media/tasks; $0.21 / second.
- [HappyHorse 1.1](https://tokely.me/models/happyhorse-1.1-i2v): ID happyhorse-1.1-i2v; Video; /v1/media/tasks; $0.16875 / second.
- [MiniMax H3](https://tokely.me/models/minimax-h3): ID minimax-h3; Video; /v1/media/tasks; $0.064 / second.
- [Claude Fable 5.1](https://tokely.me/models/claude-fable-5-1): ID claude-fable-5-1; Text; /v1/chat/completions; $7.5 input / $37.5 output per 1M tokens.
- [Claude Haiku 5.5](https://tokely.me/models/claude-haiku-5-5): ID claude-haiku-5-5; Text; /v1/chat/completions; $0.042353 input / $0.211764 output per 1M tokens.
- [Claude Opus 4.5](https://tokely.me/models/claude-opus-4-5-20251101): ID claude-opus-4-5-20251101; Text; /v1/chat/completions; $3.75 input / $18.75 output per 1M tokens.
- [DeepSeek V3.1](https://tokely.me/models/deepseek-v3-1): ID deepseek-v3-1; Text; /v1/chat/completions; $0.478335 input / $1.435005 output per 1M tokens.
- [DeepSeek V4 Pro](https://tokely.me/models/deepseek-v4-pro): ID deepseek-v4-pro; Text; /v1/chat/completions; $1.815 input / $5.445 output per 1M tokens.
- [Gemini 2.5 Flash Lite](https://tokely.me/models/gemini-2.5-flash-lite): ID gemini-2.5-flash-lite; Text; /v1/chat/completions; $0.07 input / $0.28 output per 1M tokens.
- [Gemini 3 Flash Preview](https://tokely.me/models/gemini-3-flash-preview): ID gemini-3-flash-preview; Text; /v1/chat/completions; $0.35 input / $2.1 output per 1M tokens.
- [Gemini 3.1 Pro Preview](https://tokely.me/models/gemini-3.1-pro-preview): ID gemini-3.1-pro-preview; Text; /v1/chat/completions; $1.4 input / $8.4 output per 1M tokens.
- [Gemini 3.6 Flash](https://tokely.me/models/gemini-3.6-flash): ID gemini-3.6-flash; Text; /v1/chat/completions; $0.525 input / $2.625 output per 1M tokens.
- [Gemini Embedding 001](https://tokely.me/models/gemini-embedding-001): ID gemini-embedding-001; Text; /v1/embeddings; $0.158825 input / $0.635299 output per 1M tokens.
- [GLM 5](https://tokely.me/models/glm-5): ID glm-5; Text; /v1/chat/completions; $2.0625 input / $7.425 output per 1M tokens.
- [GLM 5.1](https://tokely.me/models/glm-5.1): ID glm-5.1; Text; /v1/chat/completions; $3.85 input / $10.587504 output per 1M tokens.
- [GLM 5.2](https://tokely.me/models/glm-5.2): ID glm-5.2; Text; /v1/chat/completions; $1.925 input / $6.050001 output per 1M tokens.
- [GLM 5.3](https://tokely.me/models/glm-5.3): ID glm-5.3; Text; /v1/chat/completions; $1.85 input / $5.81 output per 1M tokens.
- [GLM 5.3 Flash](https://tokely.me/models/glm-5.3-flash): ID glm-5.3-flash; Text; /v1/chat/completions; $0.20625 input / $0.6875 output per 1M tokens.
- [GPT-3.5 Turbo](https://tokely.me/models/gpt-3.5-turbo): ID gpt-3.5-turbo; Text; /v1/chat/completions; $0.528 input / $1.584 output per 1M tokens.
- [GPT-4 Turbo](https://tokely.me/models/gpt-4-turbo): ID gpt-4-turbo; Text; /v1/chat/completions; $7 input / $21 output per 1M tokens.
- [GPT-4.1](https://tokely.me/models/gpt-4.1): ID gpt-4.1; Text; /v1/chat/completions; $1.4 input / $5.6 output per 1M tokens.
- [GPT-4.1 Mini](https://tokely.me/models/gpt-4.1-mini): ID gpt-4.1-mini; Text; /v1/chat/completions; $0.28 input / $1.12 output per 1M tokens.
- [GPT-4.1 Nano](https://tokely.me/models/gpt-4.1-nano): ID gpt-4.1-nano; Text; /v1/chat/completions; $0.07 input / $0.28 output per 1M tokens.
- [GPT-4o](https://tokely.me/models/gpt-4o): ID gpt-4o; Text; /v1/chat/completions; $1.75 input / $7 output per 1M tokens.
- [GPT-4o Mini](https://tokely.me/models/gpt-4o-mini): ID gpt-4o-mini; Text; /v1/chat/completions; $0.105 input / $0.42 output per 1M tokens.
- [GPT-5](https://tokely.me/models/gpt-5): ID gpt-5; Text; /v1/chat/completions; $0.875 input / $7 output per 1M tokens.
- [GPT-5 Mini](https://tokely.me/models/gpt-5-mini): ID gpt-5-mini; Text; /v1/responses; $0.175 input / $1.4 output per 1M tokens.
- [GPT-5.1](https://tokely.me/models/gpt-5.1): ID gpt-5.1; Text; /v1/chat/completions; $0.875 input / $7 output per 1M tokens.
- [Grok 4.20 Non Reasoning](https://tokely.me/models/grok-4-20-non-reasoning): ID grok-4-20-non-reasoning; Text; /v1/chat/completions; $0.505519 input / $1.011038 output per 1M tokens.
- [Grok 4.20 Reasoning](https://tokely.me/models/grok-4-20-reasoning): ID grok-4-20-reasoning; Text; /v1/chat/completions; $0.505519 input / $1.011038 output per 1M tokens.
- [Grok Build 0.1](https://tokely.me/models/grok-build-0.1): ID grok-build-0.1; Text; /v1/chat/completions; $0.404415 input / $0.80883 output per 1M tokens.
- [Kimi K2.5](https://tokely.me/models/kimi-k2.5): ID kimi-k2.5; Text; /v1/chat/completions; $0.825 input / $4.33125 output per 1M tokens.
- [Kimi K2.6](https://tokely.me/models/kimi-k2.6): ID kimi-k2.6; Text; /v1/chat/completions; $1.30625 input / $5.426025 output per 1M tokens.
- [Kimi K2.7 Code](https://tokely.me/models/kimi-k2.7-code): ID kimi-k2.7-code; Text; /v1/chat/completions; $1.30625 input / $5.426025 output per 1M tokens.
- [MiMo V2.6 Flash](https://tokely.me/models/mimo-v2.6-flash): ID mimo-v2.6-flash; Text; /v1/chat/completions; $0.252 input / $0.504 output per 1M tokens.
- [MiMo V2.6 Pro](https://tokely.me/models/mimo-v2.6-pro): ID mimo-v2.6-pro; Text; /v1/chat/completions; $0.522 input / $1.044 output per 1M tokens.
- [MiniMax M2.5](https://tokely.me/models/MiniMax-M2.5): ID MiniMax-M2.5; Text; /v1/chat/completions; $0.515625 input / $1.65 output per 1M tokens.
- [MiniMax M3](https://tokely.me/models/MiniMax-M3): ID MiniMax-M3; Text; /v1/chat/completions; $0.825 input / $3.3 output per 1M tokens.
- [o1](https://tokely.me/models/o1): ID o1; Text; /v1/chat/completions; $10.5 input / $42 output per 1M tokens.
- [o3](https://tokely.me/models/o3): ID o3; Text; /v1/chat/completions; $1.4 input / $5.6 output per 1M tokens.
- [o3 Mini](https://tokely.me/models/o3-mini): ID o3-mini; Text; /v1/chat/completions; $0.77 input / $3.08 output per 1M tokens.
- [o4 Mini](https://tokely.me/models/o4-mini): ID o4-mini; Text; /v1/chat/completions; $0.77 input / $3.08 output per 1M tokens.
- [Qwen Plus](https://tokely.me/models/qwen-plus): ID qwen-plus; Text; /v1/chat/completions; $1.98 input / $23.76 output per 1M tokens.
- [Qwen 3.5 122B A10B](https://tokely.me/models/qwen3.5-122b-a10b): ID qwen3.5-122b-a10b; Text; /v1/chat/completions; $0.53 input / $4.24 output per 1M tokens.
- [Qwen 3.5 27B](https://tokely.me/models/qwen3.5-27b): ID qwen3.5-27b; Text; /v1/chat/completions; $1.2375 input / $9.9 output per 1M tokens.
- [Qwen 3.5 35B A3B](https://tokely.me/models/qwen3.5-35b-a3b): ID qwen3.5-35b-a3b; Text; /v1/chat/completions; $0.33 input / $2.64 output per 1M tokens.
- [Qwen 3.5 397B A17B](https://tokely.me/models/qwen3.5-397b-a17b): ID qwen3.5-397b-a17b; Text; /v1/chat/completions; $2.0625 input / $12.375 output per 1M tokens.
- [Qwen 3.5 Plus](https://tokely.me/models/qwen3.5-plus): ID qwen3.5-plus; Text; /v1/chat/completions; $2.75 input / $16.5 output per 1M tokens.
- [Qwen 3.6 27B](https://tokely.me/models/qwen3.6-27b): ID qwen3.6-27b; Text; /v1/chat/completions; $1.2375 input / $7.425 output per 1M tokens.
- [Qwen 3.6 35B A3B](https://tokely.me/models/qwen3.6-35b-a3b): ID qwen3.6-35b-a3b; Text; /v1/chat/completions; $0.5 input / $3 output per 1M tokens.
- [Qwen 3.6 Plus](https://tokely.me/models/qwen3.6-plus): ID qwen3.6-plus; Text; /v1/chat/completions; $4.125 input / $24.75 output per 1M tokens.
- [Qwen 3.7 Max](https://tokely.me/models/qwen3.7-max): ID qwen3.7-max; Text; /v1/chat/completions; $3.4375 input / $10.3125 output per 1M tokens.
- [Qwen 3.7 Plus](https://tokely.me/models/qwen3.7-plus): ID qwen3.7-plus; Text; /v1/chat/completions; $2.475 input / $9.9 output per 1M tokens.
- [Qwen 3.8 27B](https://tokely.me/models/qwen3.8-27b): ID qwen3.8-27b; Text; /v1/chat/completions; $1.5125 input / $9.075 output per 1M tokens.
- [Qwen 3.8 Flash](https://tokely.me/models/qwen3.8-flash): ID qwen3.8-flash; Text; /v1/chat/completions; $0.309375 input / $0.969375 output per 1M tokens.
- [Qwen 3.8 Max](https://tokely.me/models/qwen3.8-max): ID qwen3.8-max; Text; /v1/chat/completions; $1.5345 input / $4.60443 output per 1M tokens.
- [Gemini 4 Argon](https://tokely.me/models/gemini-4-argon): ID gemini-4-argon; Text; /v1/chat/completions; $1.4 input / $7 output per 1M tokens.
- [Doubao Seed 2.1 Pro](https://tokely.me/models/doubao-seed-2-1-pro-260915): ID doubao-seed-2-1-pro-260915; Text; /v1/chat/completions; $0.89 input / $4.45 output per 1M tokens.
- [Kimi K3.1 Pro](https://tokely.me/models/kimi-k3.1-pro): ID kimi-k3.1-pro; Text; /v1/chat/completions; $1.6 input / $8 output per 1M tokens.
- [GPT-6.1 Terra](https://tokely.me/models/gpt-6.1-terra): ID gpt-6.1-terra; Text; /v1/chat/completions; $0.6 input / $3 output per 1M tokens.
- [Doubao Seed 2.1 Turbo](https://tokely.me/models/doubao-seed-2-1-turbo-260628): ID doubao-seed-2-1-turbo-260628; Text; /v1/chat/completions; $0.45 input / $2.25 output per 1M tokens.
- [Kling 2.6](https://tokely.me/models/kling-v2-6): ID kling-v2-6; Video; /v1/media/tasks; $0.05 / second.
- [PixVerse V5.5](https://tokely.me/models/pixverse-v5.5): ID pixverse-v5.5; Video; /v1/media/tasks; $0.036 / second.
- [Suno V5](https://tokely.me/models/suno-v5): ID suno-v5; Audio; /v1/media/tasks; $0.15 / task.
- [Text Embedding 3 Small](https://tokely.me/models/text-embedding-3-small): ID text-embedding-3-small; Embedding; /v1/embeddings; $0.01 input / $0.01 output per 1M tokens.
- [Text Embedding 3 Large](https://tokely.me/models/text-embedding-3-large): ID text-embedding-3-large; Embedding; /v1/embeddings; $0.065 input / $0.065 output per 1M tokens.
- [Text Embedding Ada 002](https://tokely.me/models/text-embedding-ada-002): ID text-embedding-ada-002; Embedding; /v1/embeddings; $0.05 input / $0.05 output per 1M tokens.
- [Jev 1.13.0](https://tokely.me/models/jev-1.13.0): ID jev-1.13.0; Decision; /v1/systemone; $0.0294 input / $0 output per 1M tokens.
- [Jev 1.14.0](https://tokely.me/models/jev-1.14.0): ID jev-1.14.0; Decision; /v1/systemone; $0.0294 input / $0 output per 1M tokens.

## Service and trust

- [Status](https://tokely.me/status): Service health and incident history; not per-model uptime certification.
- [Privacy](https://tokely.me/privacy): Current data-handling terms.
- [Terms](https://tokely.me/terms): Service and payment conditions.
- [About Tokely](https://tokely.me/about): Mission, live model and developer counts, and how a request is routed, recovered and billed.

Support: message @tokelyme on X (https://x.com/tokelyme)