| Provider | Type | What's free | The catch | Verified |
|---|---|---|---|---|
| Cloudflare Workers AI Highest free daily volume for a real side project https://api.cloudflare.com/client/v4/accounts/{account_id}/ai/v1no cardno phonecommercialOpenAI-compat textembeddingsimageaudio | Ongoing | 10,000 Neurons/day, all account plans | Resets daily at 00:00 UTC; overage on a Workers Paid plan bills at $0.011/1,000 Neurons | 2026-07-11 |
| Google Gemini API (AI Studio) The only frontier-class model with a genuine free tier https://generativelanguage.googleapis.com/v1beta/openai/no cardno phonecommercialOpenAI-compat textvisionembeddingsaudio | Ongoing | Gemini 2.5 Flash, 2.5 Flash-Lite, 2.5 Pro (limited), embeddings, TTS models | Free-tier prompts/outputs may be used by Google to improve its products when used outside the UK/CH/EEA/EU | 2026-07-11 |
| Ollama Cloud Cloud-hosted open models with the Ollama workflow, no card https://ollama.com/v1no cardno phonecommercialOpenAI-compat textvision | Ongoing | $0 Free plan: access to cloud-hosted open models (Qwen, GPT-OSS, DeepSeek, etc.) via API | First-party — Ollama hosts the cloud models. Requires an ollama.com account + API key (`ollama signin`); the free plan is for light usage, Pro ($20/mo) raises limits. No card required. | 2026-07-30 |
| OpenRouter Widest model selection behind one no-card key https://openrouter.ai/api/v1no cardno phonecommercialOpenAI-compat textvision | Ongoing | 20+ models with a :free suffix, single API across many providers | ToS (Apr 2026) prohibits resale or building a competing service on the free models; a private proxy for personal use is fine | 2026-07-11 |
| AssemblyAI Free speech-to-text plus an OpenAI-compatible LLM gateway https://llm-gateway.assemblyai.com/v1no cardno phonecommercialOpenAI-compat audiotext | Trial | $50 free credit on signup (no card) — pre-recorded & streaming speech-to-text, Speech Understanding, and an OpenAI-compatible LLM Gateway (25+ models) | One-time credit, not renewing. Catch: the default model differs between free and paid accounts — set speech_models explicitly to avoid cost jumps on upgrade. Only the LLM Gateway is OpenAI-compatible. | 2026-07-30 |
| Jina AI Free embeddings & rerank behind an OpenAI-compatible endpoint https://api.jina.ai/v1no cardno phonecommercialOpenAI-compat textimageembeddingsrerank | Trial | 10M free tokens (one-time) across all models — embeddings (v3/v4), rerankers, classifier; plus a keyless Reader (r.jina.ai) for basic use | The 10M-token balance is a one-time grant that does not replenish; the keyless Reader is genuinely ongoing. Hosted API is commercial-OK and data is not used for training. No card required. | 2026-07-30 |
| LlamaParse (LlamaCloud) Free document parsing for RAG (~10k pages/mo) no cardno phonecommercial ocrtext | Ongoing | Free plan: 10,000 credits/month (~10,000 pages in balanced parse mode at 1 credit/page) | Credit-based (1,000 credits = $1.25); premium parse modes consume more credits per page. No card required. Document parsing for RAG (LlamaIndex). | 2026-07-30 |
| OCR.space Classic free OCR (25k conversions/mo, no card) no cardno phonecommercial ocrtext | Ongoing | 25,000 conversions/month (Engine 1 & 2) plus 2,500 Engine 3 conversions/month; max 1 MB file, PDFs up to 3 pages | Free searchable-PDF output carries a watermark (raw text extraction is unrestricted); the free key needs only an email, no card. Commercial use permitted. | 2026-07-30 |
| Speechify API Free commercial TTS (50k chars/mo) no cardno phonecommercial audio | Ongoing | 50,000 characters/month TTS (hard cap) + 60 min/month voice agents | Commercial use is allowed on the free tier. No credit card required. Hard monthly cap that pauses at the limit. | 2026-07-30 |
| AI Horde Anonymous, no-signup free generation (variable speed) no cardno phone textimage | Ongoing | Free crowdsourced text & image generation; anonymous API key '0000000000' (no registration), or register to earn kudos for priority | Community-powered volunteer network — model availability and speed vary with worker supply, so it is not a fixed-SLA service. No card, no phone. Kudos never expire and cannot be sold. | 2026-07-30 |
| Deepgram Biggest free credit here, for speech (STT/TTS) no cardno phonecommercial audio | Trial | $200 free credit on signup (no card, no expiration) — Nova speech-to-text and Aura text-to-speech at pay-as-you-go rates | No card and no expiration on the credit. Data catch: the Model Improvement Program is opt-OUT — send mip_opt_out=true per request to keep your data out of training. Native REST/WebSocket API, not OpenAI-compatible. | 2026-07-30 |
| Groq Lowest-latency inference for open-weight models https://api.groq.com/openai/v1no cardphonecommercialOpenAI-compat textaudio | Ongoing | Open-weight models (Llama, Qwen, GPT-OSS) plus Whisper, no credit card required | Limits apply at the organization level, not per API key. Phone verification required at signup | 2026-07-11 |
| Mixedbread No-card embeddings & rerank credit no cardno phonecommercial textembeddingsrerankimage | Trial | Starter plan: $5 one-time credit (no card) — embeddings (mxbai-embed-large-v1), reranking, and multimodal search over PDF/image/doc/code | Free credit is usage-based and non-renewing; usage-based pricing applies afterward. Native REST API, not documented as OpenAI-compatible. No card required. | 2026-07-30 |
| Nanonets Document extraction on a $50 no-card credit no cardno phonecommercial ocrtext | Trial | $50 in free credits on signup (one-time) across data-extraction / OCR workflows | One-time signup credit, not renewing; no credit card required to start. Paid plans start at $100/month afterwards. | 2026-07-30 |
| Pollinations.ai No-signup free image generation (Flux/SD) no cardno phone imagetextaudio | Ongoing | Free hosted image models (Flux, Turbo, Stable Diffusion) via a simple GET URL; also text and audio. No signup required to start | Anonymous free images may carry a watermark (since 2025); free registration removes it via the nologo parameter. No card, no phone. Commercial use is not explicitly guaranteed in the docs. | 2026-07-30 |
| Runware Fast image models with a $2 no-card credit no cardno phonecommercial image | Trial | $2 in signup credits across first-party image models (FLUX.1/FLUX.2, Stable Diffusion 3, SDXL) on the Sonic Inference Engine — hundreds to thousands of generations | The $2 requires signup with a business email (personal/free-mail domains may be rejected); one-time, non-renewing. No watermark, no card, commercial use allowed. The OpenAI-compatible endpoint is chat-only, not for images. | 2026-07-30 |
| SambaNova Cloud Fast open models on a rate-limited free tier https://api.sambanova.ai/v1no cardcommercialOpenAI-compat text | Ongoing | Rate-limited free tier (applies when no payment method is linked) across all models | Free Tier applies when no payment method is linked to the account; SambaCloud ToS grants a commercial license (no evaluation-only clause). The previously-listed "$5 / 3 months" trial could not be re-confirmed on official pages (2026-07-30). | 2026-07-30 |
| Z.ai (Zhipu AI / GLM) A permanently-free model with no trial clock https://api.z.ai/api/paas/v4/no phonecommercialOpenAI-compat textvision | Ongoing | GLM-4.5-Flash, GLM-4.7-Flash (text), and GLM-4.6V-Flash (vision) are officially listed as $0 cost (input, cached input, and output) on a permanent basis | Terms of Use prohibit using the service to "develop, train, or improve" competing algorithms or models — otherwise general use, including commercial, isn't restricted | 2026-07-30 |
| Hume AI (Octave TTS) Expressive TTS with a commercial free tier no cardcommercial audio | Ongoing | 10,000 characters/month TTS (~10 minutes) on the Free plan | The free plan includes a commercial license; overage billed at $0.15/1,000 chars. No card stated as required to start. Not OpenAI-compatible. | 2026-07-30 |
| Cohere Free embeddings and rerank for RAG prototypes no cardno phoneeval onlyOpenAI-compat textembeddingsrerank | Ongoing | Trial (evaluation) API keys covering chat, embed and rerank | Explicitly for evaluation only — Cohere's terms prohibit production/commercial use on a trial key | 2026-07-30 |
| GitHub Models Zero extra signup if you already have a GitHub account https://models.github.ai/inferenceno cardno phoneeval onlyOpenAI-compat text | Ongoing | Included with any GitHub account via the Copilot tier | Scoped by GitHub to experimentation/prototyping, not production traffic | 2026-07-11 |
| HuggingFace Router access to many open-weight providers under one key https://router.huggingface.co/v1no cardOpenAI-compat textimageaudioembeddings | Ongoing | Free CPU Basic + ZeroGPU for Spaces; Inference Providers has a monthly credit ($0.10/mo on Free plan, $2.00/mo on PRO/Team/Enterprise) | Credits only apply with "Routed by Hugging Face" billing, not with a Custom Provider Key | 2026-07-30 |
| Nebius AI Studio Cheap open-model inference after a $1 trial https://api.tokenfactory.nebius.com/v1cardno phonecommercialOpenAI-compat text | Trial | $1 trial credit, valid for 30 days | Product renamed to "Nebius Token Factory"; a bank card is required to set up billing | 2026-07-30 |
| Alibaba Cloud (Model Studio) Qwen models with a large free token quota (Singapore region) https://dashscope-intl.aliyuncs.com/compatible-mode/v1no cardOpenAI-compat text | Trial | 1,000,000 tokens (example figure, varies by model), international/Singapore region only | Excludes batch processing, context caching, fine-tuning, and dedicated deployment | 2026-07-30 |
| Clarifai Multimodal (LLM + vision) models on a $5 credit https://api.clarifai.com/v2/ext/openai/v1no cardphoneOpenAI-compat textvisionembeddingsimage | Trial | One-time $5 credit across serverless models (GPT-OSS-120B, Claude, Llama), vision, embeddings and image generation | Catch: SMS phone verification is required to claim the $5. Credit is one-time and expires 30 days after grant; a card is required to recharge afterward. | 2026-07-30 |
| ElevenLabs Best-known TTS voices (free tier is non-commercial) no cardno phoneeval only audio | Ongoing | 10,000 credits/month shared across Text-to-Speech, Speech-to-Text and more (~10 min TTS/month) | The free tier is NON-COMMERCIAL only per the Terms of Use (a commercial license begins on paid Starter, $6/mo), and historically required attribution. API access is available on the free plan. No credit card to sign up. | 2026-07-30 |
| Fireworks AI Fast hosted open models to trial with $1 https://api.fireworks.ai/inference/v1no cardOpenAI-compat text | Trial | $1 trial credit | Default monthly spend cap of $50 for new accounts; no card needed to activate the $1 credit, card needed once it's spent | 2026-07-11 |
| NLP Cloud NLP tasks (NER, classification, summarization) on a $15 trial cardno phonecommercial text | Trial | $15 trial credit | Corrected: the official registration page does not ask for a phone number, contradicting the earlier listing | 2026-07-30 |
| Pinecone Inference Free embeddings + rerank for RAG (5M tokens/mo) no card embeddingsrerank | Ongoing | Starter (free) plan: 5M tokens/mo for embedding models (llama-text-embed-v2, multilingual-e5-large) and 500 requests/mo for the bge-reranker-v2-m3 rerank model | Free rerank limited to bge-reranker-v2-m3; overages are pay-as-you-go. Not OpenAI-compatible. No card required. | 2026-07-30 |
| Speechmatics Best free STT tier (50 hrs/mo, no card) no card audio | Ongoing | 3,000 minutes (50 hours)/month speech-to-text + 1,000,000 characters (~20 hrs)/month text-to-speech | No credit card required to start; add a card only when you exceed the free limit. Recurring monthly allowance covering both STT and TTS. Commercial-use permission not explicitly stated on the pricing page. | 2026-07-30 |
| Twelve Labs (Marengo Embed) Free multimodal (video/audio/image/text) embeddings no card embeddingsvisionaudiotext | Ongoing | Free plan: the Marengo Embed API for all input types (video, audio, image, text) at no cost, plus ~600 min indexing | No credit card for the Free plan, but indexed data expires after 90 days unless upgraded. Marengo produces multimodal embeddings. Not OpenAI-compatible. | 2026-07-30 |
| Vercel AI Gateway One key across many providers with a free monthly credit https://ai-gateway.vercel.sh/v1no cardOpenAI-compat text | Trial | Free tier with a monthly free credit covering a subset of models at lower rate limits | Free tier and its monthly-credit behaviour are confirmed in Vercel's own docs; the specific "$5/month" figure circulating in community posts is not stated there. Once you purchase credits, your account moves to the paid tier and the monthly free credit no longer applies. | 2026-07-30 |
| AI21 Labs Jamba long-context models on a $10 trial no card text | Trial | $10 trial credit, valid 3 months | Card not required for the trial credit itself, required once it expires | 2026-07-11 |
| ModelScope (API-Inference) Free daily calls to many open models (real-name verification required) https://api-inference.modelscope.cn/v1no cardeval onlyOpenAI-compat textvision | Ongoing | ~2,000 free API calls/day across open-weight models (Qwen3, DeepSeek, GLM, Llama, etc.) via API-Inference | Alibaba's model hub — a different product from Alibaba Model Studio. Requires a ModelScope account bound to an Alibaba Cloud account with real-name (ID) verification — a practical barrier for non-China users. Explicitly non-commercial ("for developers to experience"). No card. | 2026-07-30 |
| OVHcloud AI Endpoints EU-hosted, no-account access to open models https://oai.endpoints.kepler.ai.cloud.ovh.net/v1cardOpenAI-compat text | Ongoing | Several open-weight models in the catalog (e.g. Qwen3Guard) listed as $0 per token, permanently, via two access modes: anonymous (no account) and authenticated (API key tied to a Public Cloud project) | European provider (France), relevant for EU data-sovereignty/GDPR-conscious use. The authenticated tier needs a valid payment method on the project (though "Free" models themselves don't charge); anonymous access needs neither an account nor a card. A separate general $200 Public Cloud trial voucher also exists but is unrelated to this permanent free-model tier | 2026-07-11 |
| SiliconFlow Permanently-free open models, no trial clock https://api.siliconflow.com/v1phoneOpenAI-compat textembeddings | Ongoing | Several models permanently free (e.g. Qwen2.5-7B-Instruct and others) at $0 cost, plus a $1 welcome credit for paid models | Signup requires SMS phone verification. Full "real-name authentication" (needed for recharging/billing) requires a mainland China, Hong Kong/Macao, or Taiwan ID document — this may limit full access for users without one, though basic use of free models appears reachable with standard account verification | 2026-07-30 |
| Unreal Speech Largest free TTS (250k chars/mo, attribution) commercial audio | Ongoing | 250,000 characters/month TTS (~6 hours of audio) | Commercial use allowed, but free-plan users must attribute Unreal Speech with a link when publishing audio. First-party REST endpoints. Card requirement not stated. | 2026-07-30 |
| Arli AI A permanently-free (heavily rate-limited) test tier https://api.arliai.com/v1OpenAI-compat text | Ongoing | Free plan ($0): access to all text LLMs (Gemma, Qwen, etc.), capped at ~5 requests per 2-day window, 12K context, 1 request at a time | Free tier is for testing only — very restrictive. Provider advertises zero-log / no data retention. Card requirement for the free tier is not stated on the pricing page. | 2026-07-30 |
| Cerebras Fastest tokens/sec on wafer-scale hardware https://api.cerebras.ai/v1OpenAI-compat text | Ongoing | Access to all Cerebras-hosted models | Free tier includes community support (Discord) only; paid Developer tier gives "10x higher" rate limits | 2026-07-11 |
| IBM watsonx.ai (Lite plan) Governed enterprise inference with a monthly free quota cardOpenAI-compat text | Ongoing | Lite plan: 300,000 tokens/month for foundation model inference, 20 CUH/month for ML tooling, 100 pages/month of document text extraction | Lite plan doesn't support fine-tuning of foundation or custom models; 1-day idle deployment timeout. Never expires or bills while inside quota, but a payment method (with a nominal ~$1 authorization hold) is required at signup | 2026-07-11 |
| Moondream Cloud Tiny vision model, OpenAI-compatible, monthly free credit https://api.moondream.ai/v1OpenAI-compat vision | Ongoing | $5/month usage credits in every workspace (Free plan) for the Moondream vision model — caption, query (VQA), detect, point | Recurring $5/month credit; card requirement not stated on the pricing page; commercial terms not specified. OpenAI-compatible endpoint. | 2026-07-30 |
| Baseten Largest trial credit here for building something bigger https://inference.baseten.co/v1OpenAI-compat text | Trial | $30 trial credit | No documented expiration date publicly | 2026-07-11 |
| Novita AI A quick $1 to trial open-weight models OpenAI-compat text | Trial | $1 free credit on signup | Corrected from a previously listed "$0.50/year" figure, which does not appear on any official Novita domain; a separate referral program ("Give $10, Earn $10") also exists | 2026-07-11 |
| Sarvam AI Indic-language chat + speech via one OpenAI-compatible key https://api.sarvam.ai/v1OpenAI-compat textaudio | Trial | ₹100 in free credits on signup, usable across all APIs including the Sarvam-M chat/LLM API and speech (STT/TTS) | One-time signup credit shared across all APIs; card/phone requirement not stated on the docs. India-focused provider (pricing in INR), strong Indic-language models. | 2026-07-30 |
| Scaleway Generative APIs EU-hosted open models with 1M free tokens https://api.scaleway.ai/v1OpenAI-compat textaudio | Trial | 1,000,000 tokens free + 60 min Whisper transcription; billing starts at token 1,000,001 | European provider (France). Free allowance is a one-time token bucket, not time-limited | 2026-07-30 |
| Tencent Hunyuan Chinese-language LLM workloads via an OpenAI-compatible endpoint https://api.hunyuan.cloud.tencent.com/v1OpenAI-compat textvisionembeddings | Trial | 1,000,000 free tokens for Hunyuan text LLMs (hunyuan-a13b, turbos, translation & vision models), plus a separate 1,000,000-token allotment for hunyuan-embedding | Free resource package valid 1 year from activation; unused tokens expire. Tencent Cloud generally requires mainland-China real-name ID verification to activate — a practical barrier for non-China users. Commercial-use terms not stated on the free-quota page. | 2026-07-30 |
| Gladia High-accuracy async + real-time transcription to trial audio | Trial | €50 in free credits on signup for speech-to-text (~80+ hrs pre-recorded or 60+ hrs real-time at current rates) | One-time grant with no monthly reset; prepaid model — top up once the credit is consumed. Card requirement not explicitly stated on the pricing page. | 2026-07-30 |
| Modal Self-hosting open models on free monthly GPU credit text | Trial | $30/month recurring credit (Starter plan) | Corrected from a previously listed "$5-30/month" range; this is a recurring monthly credit, not a one-time trial. Modal is serverless compute you deploy models on, not a hosted model API | 2026-07-30 |
| Rime Low-latency conversational TTS to prototype audio | Trial | 3,000 free TTS minutes for every new account (Starter plan) | One-time allotment on signup; card requirement and commercial-use terms for the free minutes are not stated on the pricing page. | 2026-07-30 |
| Mistral (La Plateforme) European models to prototype against before upgrading phoneeval only text | Ongoing | "Restrictive" free tier explicitly for "try and explore" — official docs say to upgrade for "actual projects and production use" | Phone verification required to activate; free tier is opt-in for data training | 2026-07-12 |
| NVIDIA NIM Trialling NVIDIA-optimized models (evaluation only) https://integrate.api.nvidia.com/v1phoneeval onlyOpenAI-compat text | Trial | Trial credit, phone verification required | "Evaluation only, not production" per NVIDIA's own Trial Terms of Service — NVIDIA may discontinue the trial at any time with no continuity obligation | 2026-07-11 |