AI API Router Directory

OCI Generative AI (now positioned as OCI Enterprise AI) is Oracle's managed service for pretrained and custom models, agents and governance on Oracle Cloud Infrastructure, with chat, embeddings, rerank, OpenAI-compatible APIs and an OCI Responses API. Oracle bills it through the public OCI price list, where rates are itemised per model: xAI Grok and Google Gemini entries are priced per 1,000,000 tokens (with cached-input and >200K-context variants), Cohere and Meta Llama endpoints per 10,000 transactions, and private replicas or fine-tuned models on Dedicated AI units per hour. New accounts get a US$300 cloud credit valid for 30 days plus ongoing Always Free services, but Generative AI is not among the AI services Oracle lists as having a free pricing tier.

Oracle Cloud Free Tier gives a US$300 cloud credit to spend within 30 days on all OCI services in addition to Always Free services, switching to pay-as-you-go or Always Free only afterwards (oracle.com/cloud/free/); OCI Generative AI itself is not listed among Oracle AI services with a free pricing tier.

Vertex AI serves Google's Gemini models through Google Cloud's paid API (aiplatform.googleapis.com), with project-level billing and a choice between global and non-global endpoints; the generative-AI pricing URL now renders as the Agent Platform (Gemini Enterprise Agent Platform) pricing page. Published list prices per 1M tokens: Gemini 3.8 / 3.7 / 3.6 Flash are on introductory pricing of $0.75 input and $3.75 output through December 31, 2026, rising to $1.50 and $7.50 on January 1, 2027, while Gemini 3.5 Flash-Lite is $0.30 input and $2.50 output and Gemini 2.5 Flash is $0.30 and $2.50. The Vertex pricing page states no standing free monthly token allowance for Gemini; Google's free-tier documentation instead gives $300 in 90-day Free Trial credit and a separate free tier for many Gemini models under the Gemini Developer API (AI Studio).

Google Cloud Free Trial gives $300 in Welcome credit valid for 90 days for eligible new billing accounts (cloud.google.com/free/docs/free-cloud-features); the Vertex generative-AI pricing page itself lists no free token allowance but does include 5,000 free grounding queries per month across Gemini 3 models.

Amazon Bedrock is AWS's model-API layer: 18 provider brands (Anthropic, OpenAI, Meta, Mistral, DeepSeek, Google, NVIDIA, xAI, Qwen, Amazon Nova and more) behind one AWS account, priced per 1M tokens with a 50% batch discount, tiered cross-region pricing and reserved/provisioned capacity. Bedrock itself has no standing free tier — new AWS accounts get $100 in credits at sign-up plus up to $100 more over a 6-month free plan.

New AWS accounts: $100 credits at sign-up + up to $100 earned over the 6-month free plan (aws.amazon.com/free, verified 2026-09-16). Bedrock itself has no standing free tier.

Azure OpenAI (Microsoft Foundry) serves the GPT family per 1M tokens in three priced deployment scopes — Global, Data Zone (EU/US, +10%) and Regional (up to 27 regions): GPT-5.6 Sol $5/$30 (short context, Global), Terra $2/$12, Luna $0.20/$1.20, gpt-oss-120b $0.15/$0.60. Batch (−50%) and PTU reservations (from $1/PTU/hr Global) sit under the same account; new Azure accounts carry a $200/30-day trial credit.

New Azure accounts: $200 credit for 30 days (stated on the Azure OpenAI pricing page; Azure-wide, applicable to Azure OpenAI — verified 2026-09-16). No standing free tier beyond it.

Voyage AI specializes in state-of-the-art embedding models and rerankers. It provides a managed API for high-quality vector embeddings (voyage-4 family) and efficient reranking, optimized for RAG applications. Offers a highly generous free tier for developers.

No public free-tier information found in the source reference; treat free-tier availability as unverified rather than confirmed absent.

Sarvam AI is a leading Indian AI startup focused on large-scale models for Indic languages. Their API platform provides access to the Sarvam series models for text and voice, targeting applications in the Indian market with high-performance regional language support.

No public free-tier information found in the source reference; treat free-tier availability as unverified rather than confirmed absent.

Baichuan's open platform sells Baichuan4-family and M-series models with billing quoted per 1,000 tokens and 32K context on most SKUs: Baichuan4-Turbo at 0.015 CNY per 1K tokens with input and output combined, Baichuan4-Air at 0.00098 CNY per 1K, Baichuan4 at 0.1 CNY per 1K, and the medical M3-Plus and M3 models split input and output (0.005 / 0.009 and 0.01 / 0.03 CNY per 1K tokens). Search enhancement is metered separately at 0.03 CNY per call, and Baichuan2-53B has a documented off-peak window at half price. New accounts are described as receiving an 80 CNY gift balance valid for 3 months.

Official prices page states new accounts registered after the afternoon of 22 May 2024 receive an 80 CNY gift balance (equivalent to 10M Baichuan2-Turbo tokens) valid 3 months; converted at roughly 7.1 CNY/USD, so the USD figure is an estimate of a CNY grant.

DigitalOcean's AI stack is GPU Droplets plus a managed Inference product (model catalog, serverless and dedicated inference, agents and knowledge bases) rather than one hosted LLM endpoint. GPU Droplets bill per second from $0.76/GPU/hour (RTX 4000 Ada) up to $11.19/GPU/hour spot or $7.94 reserved for HGX B300, and serverless inference is prepaid only, charging published per-token rates such as Llama 4 Maverick at $0.20 input / $0.696 output per 1M tokens and gpt-oss-120b at $0.055 / $0.385. New accounts can receive a $5 credit valid for 90 days.

A $5 new-account credit applied to usage and expiring after 90 days is described in the 'How does the $5 free credit work?' FAQ on the GPU Droplets pricing page; no larger AI-specific credit is stated on the official pages fetched.

v0 by Vercel is an AI-powered full-stack web application builder that generates React components and backend logic. Usage is metered on input and output tokens which convert to credits. It provides access to various models including v0 Mini, Pro, and Max tiers.

No public free-tier information found (researched 2026-08-29).

Dify is an open-source LLM app development platform. Its cloud service (Dify Cloud) provides hosted model access via an API, with credits consumable across OpenAI, Anthropic, Gemini, and others. Details were checked against the provider's own page at dify.ai.

200 message credits on Sandbox plan; value varies by model (e.g. GPT-4o = 10 credits).

PiAPI offers a model aggregation platform with native API formats. It provides subscription tiers that include bonus credits and pay-as-you-go usage. New users receive a $0.50 starting credit. Details were checked against the provider's own page at piapi.ai.

No public free-tier information found (researched 2026-08-29).

Stability AI sells hosted generative media through a credit-based developer API plus a separate subscription studio. The developer pricing page states API usage is credit-based with 1 credit = $0.01, and lists per-operation credits: Stable Image Ultra 8 credits, Stable Diffusion 3.5 Large 6.5, SD 3.5 Large Turbo 4, SD 3.5 Medium 3.5, SD 3.5 Flash 2.5, Stable Image Core 3, SDXL 1.0 from 0.9, Creative Upscaler 60, Stable Audio 3.0 26 and Stable Fast 3D 10. New developer accounts get 25 free credits, i.e. about $0.25 of usage. The separate Brand Studio product free tier is stated at 1000 credits, then $19/month for 2000 credits or $50/month for 5000 credits, with unused credits not rolling over.

25 free credits for new accounts, stated on platform.stability.ai/pricing alongside the fixed rate "1 credit = $0.01" - equivalent to about $0.25.

OpenAI Codex is the coding agent that runs across ChatGPT, an IDE extension, the CLI and cloud environments on a single ChatGPT account, with cloud worktrees, scheduled background tasks, Skills and code review. The Codex pricing documentation lists Free at $0/month for "quick coding tasks", Go at $8/month, Plus at $20/month, Pro from $100/month, Business at $20/user/month billed annually ($25 monthly, minimum 2 users) and Enterprise/Edu by contact, plus an API-key route billed at API rates. Usage is measured as messages per five-hour period that vary by model: on Plus, the documented ranges run from 250-2,000 messages for GPT-5.6 Luna down to 5-45 for GPT-6 Astra, and OpenAI states these are estimates rather than fixed limits.

Codex Free is $0/month, described as "Explore Codex capabilities on quick coding tasks"; no fixed free message count is published (Codex pricing docs, checked 2026-09-16).

360 AI (360智脑) is a Chinese LLM API provider by 360 Group, offering conversational models with support for multiple file formats and LlamaIndex compatibility. Access is managed through a dashboard-generated API key, primarily serving enterprise and developer needs in China.

No public free-tier information found in the source reference; treat free-tier availability as unverified rather than confirmed absent.

AI/ML API is a major AI model aggregator providing unified access to 1000+ models from OpenAI, Anthropic, Google, DeepSeek, and more. It offers a single API key for LLM, image, voice, and video models with a focus on production-ready pay-as-you-go access.

No public free-tier information found in the source reference; treat free-tier availability as unverified rather than confirmed absent.

AINative Studio specializes in AI agent infrastructure, providing multimodal APIs, persistent agent memory, and MCP server hosting. It offers tiered subscription plans that include 'ZeroTime' credits for agentic workflows and specialized model access.

No public free-tier information found in the source reference; treat free-tier availability as unverified rather than confirmed absent.

Ant Ling (by inclusionAI) provides an OpenAI-compatible API for its proprietary Ling model family. It supports high-speed inference and is designed to integrate seamlessly with standard SDKs by switching the base URL, serving both developer and enterprise needs.

No public free-tier information found in the source reference; treat free-tier availability as unverified rather than confirmed absent.

Arcee AI provides a domain-specific model platform (Arcee Conductor) and inference API for open-weight models. It uses a native API format but Conductor provides a unified interface. Access is pay-as-you-go with no added premium on 3rd party model inference costs.

No public free-tier information found in the source reference; treat free-tier availability as unverified rather than confirmed absent.

Auriko is an AI API router and aggregator that provides a unified interface to multiple LLM providers. It offers both platform-managed and Bring Your Own Key (BYOK) modes. It publishes high rate limits for both modes.

No public free-tier information found in the source reference; treat free-tier availability as unverified rather than confirmed absent.

Azure AI Foundry (formerly Azure AI Studio) provides access to a vast catalog of models including OpenAI, Anthropic, and Llama. It features a unified API surface and enterprise-grade management. Pricing is unified under Azure's pay-as-you-go model.

No public free-tier information found in the source reference; treat free-tier availability as unverified rather than confirmed absent.

Baidu ERNIE models are served through the Baidu Qianfan model service and Agent development platform on Baidu Cloud, alongside third-party models such as DeepSeek, Kimi and Qwen. From 24 October 2025, users who open Qianfan and accept the user agreement automatically receive new-user free token quota: 17 listed preset models each grant 1,000,000 tokens valid for 3 months, including ERNIE-4.5-Turbo-128K, ERNIE-4.5-Turbo-32K, ERNIE-4.5-Turbo-VL, ERNIE-X1-Turbo-32K, bge-large-en and bge-large-zh. Free quota offsets online inference of preset models only, not batch inference, and verified unit prices on the billing page include embeddings at 0.0005 CNY per thousand tokens, image generation from 0.05 CNY per image, and web search augmentation at 0.004 CNY per triggered search.

Official page states a new-user grant of 1,000,000 tokens on each of 17 preset models, valid 3 months (cloud.baidu.com Qianfan new-user free quota doc); no USD value is given and the ERNIE text-model unit-price table did not render on the official billing page, so no conversion was made (checked 2026-09-16).

Baidu Qianfan is the official Model-as-a-Service (MaaS) platform of Baidu Intelligent Cloud. It serves the ERNIE (Wenxin Yiyan) series models and other third-party models via an OpenAI-compatible v2 API, offering high-performance inference for enterprise and individual developers.

No public free-tier information found (researched 2026-08-29).

BytePlus ModelArk is ByteDance's international enterprise AI platform providing access to the Doubao model family. It features a tiered pricing structure and integrates with BytePlus's global cloud ecosystem.

No public free-tier information found (researched 2026-08-29).

Hyper (by Charm) is an AI inference solution purpose-built for coding. It manages its own infrastructure to optimize open-source coding models. It provides 100 free Hypercredits monthly on signup. Details were checked against the provider's own page at hyper.charm.land.

No public free-tier information found in the source reference; treat free-tier availability as unverified rather than confirmed absent.