AI API Router Directory

Cerebras Inference serves open-weight models on wafer-scale CS-4 hardware — gpt-oss-120b at ~3,000 tokens/s for $0.35/$0.75 per 1M in/out, Qwen 3.8 27B at $0.99/$1.49. The self-serve developer tier is pay-as-you-go with $5 of free credit to start; the enterprise tier (production capacity, priority limits, fine-tuning) is quote-based. Also available via AWS Marketplace, OpenRouter, Hugging Face and Vercel.

Self-serve developer tier includes $5 free credit to start (pricing page, checked 2026-09-16). Replaces an older $30 figure that no longer matches the official page.

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.

EasyMax (Easy API) is a multi-model aggregation gateway compatible with OpenAI and Claude API protocols. It focuses on stability and pricing, supporting GPT, Gemini, and image generation models. Details were checked against the provider's own page at easymax.ai.

Relay with ~5-8 free model entries on /api/pricing; V2EX promo grants $5 trial credit for forum reply (one-time, promotional). Continuous free quota limited; low-cost paid top-ups.

Firecrawl: web-to-LLM-context platform — scrape, crawl, map, search, extract, and browser Interact APIs. Credits-based: 1 credit = 1 page (scrape/crawl/map); search costs 2 credits per 10 results. Free 1,000 credits/month (500 searches) with no card; paid subs from $16-19/mo with pay-as-you-go overage in $5 increments (paid plans only).

Free 1,000 credits/mo ≈ 500 searches (2 credits per 10 results) or 1,000 pages; Firecrawl publishes no per-search dollar price for the free allocation, so value left null.

Flatkey.ai is a commercial AI API gateway aggregating over 200 official models under a single API key. It uses a prepaid pay-as-you-go model, claiming to be significantly cheaper than official pricing by leveraging high-volume enterprise agreements.

Search reports a $5 initial credit for testing.

FreeModel is a frontier model API router that provides a single, unified endpoint for top-tier models like GPT-5 and Claude Code. It features intelligent routing to optimize for output quality and cost, supporting OpenAI-compatible SDKs for seamless developer integration.

Official site: free plan '永久免费·无需信用卡' (forever free, no CC) incl. '注册即送 $5 API 额度' ($5 signup credit) + FRE-5.4/5.5 models. Community reported a $300 signup promo (May 2026, likely expired). OpenAI+Anthropic compatible.

IAMHC (Xinjiang Huancheng Wangan) is a public-welfare AI API gateway providing unified access to over 199 mainstream models. It features a rate-limited free tier and emphasizes compliance with built-in moderation for developers in the China region.

Free tier: small-parameter models only, ~1500 calls per 5h (some sources say ~800), unguaranteed and test-only; paid pub/VIP/SVIP tiers. Rough ≈$5/mo equivalent; reliability low.

InternLM (Shanghai AI Laboratory) provides an open platform with an OpenAI-compatible API. It offers a monthly free token allowance for its InternVL and long-thinking reasoning models. Details were checked against the provider's own page at internlm.intern-ai.org.cn.

Community users: ~1M input + 3M output tokens free per month (~10 RPM, keys valid 6 months) ≈ $5/mo at $0.5 in / $1.5 out per 1M; monthly usage visible in console under API Usage.

Nscale's Serverless Inference is an OpenAI-compatible endpoint service for open models - the catalogue pages list Meta Llama 4 Scout, Alibaba Qwen, DeepSeek, OpenAI GPT OSS 20B/120B, Mistral 8x22B Instruct, FLUX.1 [schnell] and Stable Diffusion XL Base 1.0 - with token-based pay-as-you-go billing. Per Nscale's own launch materials, pricing is based on input and output tokens for text and vision models, while image generation is priced on output image dimensions. Per-token rates are not published on a static official page: the docs direct users to the console's AI Services to Models list or the /v1/models API endpoint for pricing and context length. Every new user can claim $5 of free credit, and Nscale states the platform does not log or train on request or response content.

Every new user can claim $5 of free credit, stated in Nscale's Serverless Inference launch post (2 April 2025) and in its announcement that pricing is based on input and output tokens for text and vision models.

OpenInference - open-source LLM relay with a single OpenAI-compatible endpoint (Anthropic-format root also available at https://api.openference.com/), API-key auth, automatic provider failover and a zero-data-retention option; free to start, paid plans from USD 1/mo; documented per-model on-demand USD/1M pricing on openference.com/models.

Free plan: 3-day trial with open-source models — no credit card required

Pollinations.ai is a Berlin-based open-source platform providing a signup-free, OpenAI-compatible API for text, image, and audio generation. It aggregates multiple open-weights models (DeepSeek, Qwen, Mistral) and offers them for free to foster creative AI development.

Basic use needs no signup and no API key; new Pollen credit system gives ~1.5 Pollen/week free, with some models/higher limits requiring an account or paid tier. Pollen has no published USD rate - not quantifiable.

Vercel AI Gateway is a unified API gateway that allows routing to hundreds of models from various providers through a single OpenAI-compatible endpoint. It provides observability, budgets, and zero-data-retention (ZDR) options with no token markup over upstream list prices.

Each team account gets $5 of AI Gateway credit every 30 days to try any model; after your first payment you become a paid customer and stop receiving the free credit. Official figure.

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).

DoCode is a relay specifically optimized for AI coding workflows (Claude Code, Cursor, Codex). It uses heavy caching and model routing to lower costs, with VIP tiers for further discounts. Authentication via sk- key.

$200 trial credits on signup; estimated real value based on ¥1:$50 ratio.

AnyAPI is a commercial multi-model API aggregation gateway providing access to 400+ models via a single key. It offers a generous free tier for developers and scales to high-volume enterprise plans, supporting models from OpenAI, Anthropic, and open-weights families.

Free tier $0/mo: 100K tokens/day (~100 credits/day, 1 credit≈1K tokens), ~20 RPM / 200 RPD, free and basic models only, no card. ≈$3/mo blended at $1 per 1M tokens.

Dahl Inference is a high-performance platform offering decentralized access to open models like DeepSeek and MiniMax. It features an OpenAI-compatible API and a generous signup promotion of 100 million free tokens.

Value based on 100M free tokens at an estimated rate of $0.03/1M.

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).

Qiniu Cloud AI LLM Inference is a managed MaaS platform that provides a unified, compliant API for over 50 mainstream LLMs. It is compatible with both OpenAI and Anthropic protocols, offering new users substantial free tokens and flexible resource packages for production scaling.

First API key activates a shared 3M-token free resource pack (~$3 at $1/1M; auto-granted ~1h after first billable usage), then PAYG; several open-source models are permanently fully free (bonus, unquantified).

VSLLM (Weiyun Model Open Platform) is a commercial AI API relay gateway that aggregates multiple large models. It is OpenAI-compatible and features a daily auto-refreshing free quota for new sign-ups. Claims to offer highly competitive, transparent pricing based on actual token usage.

Free subscription auto-refreshes daily without claiming: ~20 API calls + ~100K tokens/day (non-cumulative, reset next day) ~ $3/mo at $1/1M tokens; also GitHub-star and daily check-in reward activities.

Serper: fast (1-2s) Google SERP API — web, news, images, places, scholar, patents, videos, shopping, autocomplete as structured JSON. Credit-based prepaid billing: 2,500 free credits on signup, $1.00/1k base dropping to $0.30/1k at volume; credits valid 6 months; failed requests not billed. (Official pricing page is 404 — figures corroborated across multiple independent 2026 sources.)

2,500 free signup credits × $1.00/1k base rate ≈ $2.50 (one-time, credits expire in 6 months).

SiliconFlow (SiliconCloud) is a premier Chinese AI model platform offering high-performance inference for a wide range of open-source models (Qwen, Llama, DeepSeek). It is notable for providing free, permanent API access to models with parameters under 9B, alongside competitive pay-as-you-go rates for larger flagship

Signup grants ~CNY 14 (≈$2.08) credit, plus ~CNY 16 (≈$2.38) voucher after real-name verification; many <=9B open-source models (Qwen2.5-7B, Llama3.1-8B, etc.) are permanently free; direct access within mainland China. Since 2026-05-15, unverified users are capped at 10 RPM and delinquent/unverified accounts are cut off.

platform.kimi.com is the Kimi (Moonshot AI) open platform; the China-facing site quotes CNY while the published USD price list is at platform.moonshot.ai/docs/pricing/chat. Kimi K3 is listed at $3.00 per 1M input tokens (cache miss), $0.30 cached input and $15.00 output with a 1,048,576-token context; K2.7 Code is $0.95 / $0.19 / $4.00 and K2.6 is $0.95 / $0.16 / $4.00, both with 262,144-token contexts. The API is paid prepaid usage rather than subscription, and the only free grant documented is an exchange voucher of 15 CNY issued after identity verification, which the docs state cannot be spent on Kimi K3.

Official docs state a 15 CNY voucher is granted after personal or enterprise identity verification (platform.kimi.com/docs/guide/account-and-payments), converted at roughly 7.1 CNY/USD; the same page states the voucher cannot be used on Kimi K3.

platform.moonshot.cn is the China-domain entry point for Moonshot AI's Kimi platform, and it serves the same documentation tree as platform.kimi.com - the fetched pricing page points its documentation index at platform.kimi.com/docs/llms.txt, so model IDs and parameters are identical across the two domains. Prices are quoted in CNY on the China-facing site, which lists K3 at 20.00 CNY input / 2.00 CNY cached input / 100.00 CNY output per 1M tokens, K2.7 Code at 6.50 / 1.30 / 27.00 and K2.6 at 6.50 / 1.10 / 27.00, while the international list quotes K3 at $3.00 / $0.30 / $15.00 per 1M tokens. As on the parent platform, billing is prepaid token usage and the only documented free grant is the 15 CNY verification voucher, which excludes K3.

Same account system as platform.kimi.com: a 15 CNY voucher is granted after identity verification (platform.kimi.com/docs/guide/account-and-payments), converted at roughly 7.1 CNY/USD, and cannot be used on Kimi K3.

Cartesia provides Sonic text to speech, Ink speech to text and Managed Agents voice agents, priced through credits and prepaid agent minutes with unlimited workspace seats on every plan. The Free plan costs $0 and includes 20,000 credits per month (about 27 minutes of Sonic-3.6 text to speech) plus $1 of prepaid agent minutes; Pro is $5 per month with 100,000 credits, Startup is $49 per month with 1.25M credits, and Scale is $299 per month with 8M credits. Agent calls bill at $0.06 per minute on every tier, and telephony is $0.014 per minute when using a Cartesia-provided phone number.

Free plan grants 20,000 credits per month (about 27 minutes of Sonic-3.6 text to speech) plus $1 of prepaid agent minutes; the credits are worth about $1.00 at the Pro plan rate of $5 per 100,000 credits, giving roughly $2.00 combined, an estimate because Cartesia does not price the free allowance in USD (cartesia.ai/pricing, checked 2026-09-16).