induwara.lk
induwara.lkDevelopers · AI

AI Free Tier Comparison — Free LLM API Limits

Compare the free tier of every major LLM API — free requests per day, rate limits, credit-card requirement, vision support and whether your prompts are used for training. Enter your daily volume and get the best no-cost option. 8 providers, sources cited, no signup.

By Induwara AshinsanaUpdated Jun 14, 2026
Free LLM API limits, side by side8 providers

Your usage

How many API calls your app makes in a day. One chat reply = one request.

Filters
Quick volumes
Best free option for your usage
Google GeminiGemini 2.0 Flash (free tier)

1,500 free requests/day, no credit card, supports image input — covers your 200/day with headroom.

Get a key from Google Gemini

All free tiers (8 of 8)

ProviderStatusFree RPDDocs
Mistral (La Plateforme)
Mistral / Pixtral (Free Experiment tier)
No fixed daily cap—Docs
Google Gemini
Gemini 2.0 Flash (free tier)
Covered1,500Docs
Groq
Llama 3.3 70B Versatile (free tier)
Covered1,000Docs
OpenRouter
Free model variants (":free" suffix)
Not covered50Docs
Cohere
Command / Aya Vision (trial key)
Not covered33Docs
Hugging Face
Inference Providers (serverless router)
Credit-based — no fixed daily cap—Docs
OpenAI
GPT-4o mini / GPT-4.1 (trial + data-sharing tokens)
Trial credits only—Docs
Anthropic (Claude)
Claude (trial credits)
Trial credits only—Docs
No API key, no proxy, no logging.

This page compares documented limits only — it never calls a provider API. Numbers were last verified on 2026-06-14; free tiers move often, so confirm at each provider's docs (linked per row and in the Sources section below) before committing.

How it works

This is a curated reference table with a thin coverage check on top — no opaque scoring. Each provider row records its documented free-tier limits, and the tool answers one question: which free tier covers your daily request volume without breaking your constraints?

  1. Coverage check. For a provider with a documented daily cap, your usage is covered when its free requests/day (RPD) is at least your entered volume, borderline when it covers at least 80% of it, and not covered otherwise.
  2. Rate-based tiers. Some providers (Mistral) publish no fixed daily cap — just a per-second rate. Their effective daily ceiling is requests/minute × 1,440 (minutes in a day). They're shown as “no fixed daily cap” and cover any volume up to that ceiling.
  3. Filters.“Needs vision” keeps only providers whose free model accepts image input; “No credit card” keeps only tiers you can start without a payment method. Both come from each provider's sign-up docs.
  4. Best-option pick.Among tiers that pass your filters and cover your volume, the tool recommends the one with the highest free RPD (tie-broken by tokens/minute). A fixed daily allowance beats a rate-based tier when it covers you, because it's the clearer promise. If nothing fully covers your volume, it surfaces the closest free option and flags that it will rate-limit.
  5. No live API calls.Every number is static and cited. The page never contacts a provider API — it only compares documented limits, so there's nothing to log and no key to leak.

As a credibility cross-check, every row is validated so its documented daily cap never exceeds what its per-minute limit physically allows (RPM × 1,440). Because free tiers shift often, the LAST_VERIFIEDdate is shown on the tool and each row links to the provider's live rate-limit docs. For paid estimates beyond the free tier, see the AI Chatbot Cost Calculator and the AI Model Comparison.

Worked examples

Hobby chatbot, no credit card

300 requests/day · vision off · no credit card on

  1. Filter to no-card providers: Gemini, Groq, Mistral, OpenRouter, Cohere, Hugging Face.
  2. Coverage by fixed RPD: Gemini 1,500 ≥ 300 → covered; Groq 1,000 ≥ 300 → covered.
  3. OpenAI and Anthropic need a card → filtered out.
  4. Highest covering free RPD wins: Gemini at 1,500/day.
  5. Result: Gemini 2.0 Flash — 1,500 free requests/day, no card, covers 300/day with headroom.

Vision app, higher volume

2,000 requests/day · vision on · credit card allowed

  1. Filter to vision providers: Gemini, Mistral, OpenRouter, Cohere, Hugging Face, OpenAI, Anthropic.
  2. Fixed-RPD tiers: Gemini 1,500 < 2,000 → not covered (no fixed-cap tier covers 2,000).
  3. Rate-based check: Mistral sustains 1 req/sec ≈ 86,400/day ≥ 2,000 → covered.
  4. Result: Mistral (Pixtral) — no fixed daily cap, covers 2,000/day with vision.
  5. For quality and latency at this volume, compare the paid tiers in the cost calculators above.

Edge case — volume beyond any free tier

100,000 requests/day · vision on · no credit card on

  1. Filter to no-card vision providers: Gemini, Mistral, OpenRouter, Cohere, Hugging Face.
  2. No fixed-RPD tier covers 100,000/day (Gemini caps at 1,500).
  3. Mistral's sustained ceiling 86,400/day < 100,000 → not fully covered either.
  4. Tool picks the closest by capacity: Mistral (86,400/day), flagged 'will rate-limit'.
  5. Takeaway: above ~86,000 requests/day, no free tier keeps up — budget for a paid plan.

Frequently asked questions

Sources & references

Every figure on this page was cross-checked against these provider docs on 2026-06-14. Free tiers change frequently; this page is reviewed monthly and after any major provider policy change.

Related tools

LiveAI

AI Prompt Library

48 structured, source-cited prompt templates for ChatGPT, Claude, Gemini, and Llama. Categories cover writing, coding, analysis, learning, marketing, and productivity. Filter, fill placeholders, copy in one click — no signup, no API call.

Open tool
LiveAI

AI Chatbot Cost Calculator

Estimate the monthly API cost of a multi-turn AI chatbot across Claude, GPT, and Gemini. Models the quadratic context re-sending that single-call calculators miss, with and without prompt caching, in USD and LKR.

Open tool
LiveAI

AI Knowledge Cutoff

Look up any major LLM's training-data knowledge cutoff and release date, straight from the provider's model card, with how stale its knowledge is today.

Open tool
Rate this tool
Be the first to rate

Comments & feedback

Spotted a bug or want an improvement? Tell us — our team reviews every comment, and good ideas get built. Comments are public and anonymous.

Spotted a free tier that changed, or a provider I missed?

Email me at [email protected] — most fixes ship within 24 hours.