Every AI company offers a free tier. Most developer guides list them without mentioning the actual limits. Here's the honest breakdown — what you get for free, when you'll hit the wall, and what the upgrade costs.
Last verified: 15 August 2026.
I re-check this page against each vendor's own docs, not against other people's blog posts. In the big August sweep every single row moved: two providers stopped publishing their free-tier numbers entirely, one killed its free trial outright, and one turned its free tier into a ten-cent credit pool. The 15 August re-check caught several more — most notably that Google's current Gemini 3.6 Flash price is half what this page was quoting (details below) and that xAI shipped a new flagship, along with corrections to how Groq's plans are named and what Replicate's free access actually is. Where a vendor no longer publishes a number, I say so instead of quoting the old one.
The Complete Free Tier Comparison
| Provider | Free access | Rate limit | Volume cap | Best model available | Source (checked 2026-08-15) |
|---|---|---|---|---|---|
| Google Gemini | Free tier, $0 in / $0 out | No longer published | No longer published | Gemini 2.5 Flash, Gemini 3.6 Flash | rate limits |
| Groq | Free plan, $0 | 30 RPM, 12K tokens/min | 1,000 req/day, 100K tokens/day | Llama 3.3 70B | rate limits |
| Anthropic Claude | "A small amount" of credit, no figure given | No published free-tier RPM | Start tier: $500/month spend cap | Claude Sonnet 5 | pricing |
| OpenAI | Free tier exists, but not for GPT-4.1 mini | 500 RPM at Tier 1, after $5 paid | 10,000 req/day, 200K tokens/min | GPT-4.1 mini | gpt-4.1-mini |
| Together AI | None. No free trials | n/a | $5 minimum purchase to use at all | Llama 3.3 70B | billing docs |
| HuggingFace | $0.10/month credit | n/a | $0.10/month | Mostly CPU-scale models | pricing |
| Replicate | Select models free to run, then billing required | 6 req/min (no card); 600 req/min otherwise | Pay per prediction | Stable Diffusion, Flux | rate limits |
| xAI / SpaceXAI (Grok) | Signup credit not confirmable | 150 RPS (T0, Grok 4.6 and 4.5) | 50M tokens/min (T0) | Grok 4.6 | rate limits |
| Mistral | "Experiment" tier no longer confirmable | No longer published | No longer published | Mistral Small 4 | API pricing |
| Cohere | Free trial key | 20 req/min (Chat API) | 1,000 API calls/month | Command A+ (trial-key access unconfirmed) | rate limits |
Tier 1: Best Free Tiers (Actually Usable)
Google Gemini API
Still the most generous free tier I've used. It's also the one that got harder to write about honestly.
Google no longer publishes free-tier RPM, TPM or requests-per-day in a per-model table. The rate limits page now says your limits "depend on a variety of factors (such as your usage tier) and can be viewed in Google AI Studio," and the link to view them requires a Google login. So the "15 RPM, 1M tokens/day" figure this article carried in April is no longer confirmable from a primary source, and I've pulled it. Check your own console. That's the only number that applies to your account anyway.
import google.generativeai as genai
genai.configure(api_key="YOUR_API_KEY")
model = genai.GenerativeModel("gemini-2.5-flash")
response = model.generate_content("Explain MCP servers in 2 sentences")
print(response.text)
Free limit: $0 input and $0 output on the free tier. The request ceiling is console-only now.
Paid tier: $0.30 per 1M input tokens, $2.50 per 1M output, standard tier
Best for: Prototyping, high-volume batch processing, long-context tasks
Correction worth flagging: this page previously said Gemini 2.5 Flash costs $0.15/1M input. That's the Batch tier price, not standard. Standard is double it. If you budgeted off the old number, you're off by 2x.
Gemini 2.5 Flash has no announced shutdown date, so it's safe to build on. Gemini 3.6 Flash was released 21 July 2026 and is also $0 on the free tier. Its paid price is date-banded on Google's own pricing page: $0.75 per 1M input and $3.75 per 1M output through 31 December 2026, doubling to $1.50/$7.50 from 1 January 2027. Correction (15 August): this page previously printed $1.50/$7.50 as today's price — that's the 2027 price, twice what you'd actually pay right now. Either way it's well above 2.5 Flash, so for most free-tier work 2.5 Flash is still the sensible default.
Groq — Fastest Inference
Groq runs open-source models on custom LPU hardware, and llama-3.3-70b-versatile is still an active production model there. A naming correction from earlier versions of this page: Groq's console lists two separate plans — a Free plan at $0 and a paid, pay-per-token Developer plan. The numbers below are the Free plan's, and that Free plan is the one entry in this table that's both genuinely free and genuinely usable. The Developer plan is what you upgrade to for higher limits plus Batch and Flex processing.
from groq import Groq
client = Groq(api_key="YOUR_API_KEY")
response = client.chat.completions.create(
model="llama-3.3-70b-versatile",
messages=[{"role": "user", "content": "Write a Python function to validate email"}]
)
print(response.choices[0].message.content)
Free limit (Free plan): 30 RPM, 12,000 tokens/min, 1,000 requests/day, 100,000 tokens/day
Paid tier: Developer plan, pay per token
Best for: Real-time applications, chatbots, any use case where latency matters
Two corrections here. The tokens-per-minute figure is 12,000, not the 15,000 this page said before. And the daily ceilings are real: 1,000 requests and 100,000 tokens per day. Nobody notices those until they run a batch job, and then they notice hard. 100K tokens/day is maybe forty decent-sized prompts with long answers. Budget accordingly.
xAI / SpaceXAI (Grok API)
The widely-repeated claim is a $25 signup credit plus $150/month for opting into data sharing. I could not confirm either figure. I checked xAI's billing FAQ and its console billing docs directly, most recently on 15 August. Neither page mentions a signup credit or a data-sharing credit. The only $25 I found was a minimum auto-top-up purchase amount, which I suspect is where the rumour started. The promo may well exist. But xAI doesn't document it, and I'm not stating it as fact because four affiliate sites copied each other.
What xAI does publish is rate limits in requests per second. The entry tier (T0) on Grok 4.6 and Grok 4.5 is 150 RPS and 50M tokens/min. xAI derives that per-second figure from your per-minute budget — its docs state the per-second limit is RPM divided by 60, so you can't spend a whole minute's requests in one burst.
A naming note: xAI was absorbed into SpaceX and rebranded SpaceXAI on 6 July 2026, though the API docs, the docs.x.ai domain and support@x.ai all still say xAI. I use both names here so this page turns up whichever one you searched. Also: "Grok 3 mini" is gone. The current flagship is Grok 4.6, which appeared on xAI's models page between my 10 August and 15 August checks — the docs call it "the most intelligent and fastest model," at $2.00/$6.00 per 1M tokens under 200K context, the same price as Grok 4.5. The rest of the lineup is Grok 4.5, Grok 4.3, the Grok 4.20 family, and Grok Build 0.1. Grok Build 0.1 is the cheapest at $1.00/$2.00 per 1M tokens under 200K context.
Tier 2: Credits That Aren't What You Think
Anthropic Claude API
Claude is still what I reach for when the code has to be right. The free-tier story changed shape completely, though.
import anthropic
client = anthropic.Anthropic(api_key="YOUR_API_KEY")
message = client.messages.create(
model="claude-sonnet-5",
max_tokens=1024,
messages=[{"role": "user", "content": "Write a React hook for debounced search"}]
)
print(message.content[0].text)
The model ID matters. This page previously used claude-sonnet-4-20250514, which is now listed as retired except on Bedrock and Google Cloud — it will error against the first-party API. And the table said "Claude 3.5 Sonnet," which no longer appears on Anthropic's current or legacy model tables — it survives only in the deprecation history, retired 28 October 2025. The correct current string is claude-sonnet-5. If you copied the old snippet, that's your bug.
Price, and the date on it just disappeared: Claude Sonnet 5 is $2 per MTok input and $10 per MTok output — and as of August 2026 that's no longer introductory. Anthropic's pricing page now says it plainly: "The $2/$10 per million input/output token pricing for Claude Sonnet 5, announced at launch as introductory pricing through August 31, 2026, is now the standard price. The previously scheduled increase to $3/$15 per million input/output tokens on September 1, 2026 will not occur." If you budgeted for the September jump this page previously warned about, you can unbudget it — re-checked against that page on 15 August 2026.
Free credit: Anthropic's pricing FAQ now says only that "new users receive a small amount of free credits to test the API." No dollar figure. The old $5 number is still repeated everywhere, but it's not on Anthropic's page anymore, so I'm not printing it as fact.
Rate limits: there is no numbered free tier. Anthropic publishes usage tiers keyed to a monthly spend cap. The lowest documented one (Start) gives Sonnet 5 1,000 RPM, 2,000,000 input tokens/min and 400,000 output tokens/min against a $500/month cap. New organisations may land in an "Evaluation tier" with lower limits that Anthropic doesn't publish. The "5 RPM free tier" framing this page used before doesn't describe anything that exists.
If you want Claude for daily coding, the API isn't the play — Claude Code is included in Claude Pro, which is $17/month billed annually or $20/month billed monthly, and that is far more cost-effective than metering an agentic session. How the two compare, and I've written up whether the Claude API is free in more detail.
OpenAI API
This row was backwards before, and the direction of the error matters.
GPT-4.1 mini has no Free-tier access. None. OpenAI's model page starts its rate-limit table at Tier 1, and Tier 1 requires $5 paid, not $5 given. Once you've paid it, you get 500 RPM, 10,000 requests/day and 200,000 tokens/min. That's not 3 RPM. It's roughly 167 times more than this page claimed.
There is a genuine no-payment Free tier at OpenAI, restricted to allowed geographies with a $100/month usage limit. GPT-4.1 mini just isn't on it.
from openai import OpenAI
client = OpenAI(api_key="YOUR_API_KEY")
response = client.chat.completions.create(
model="gpt-4.1-mini",
messages=[{"role": "user", "content": "Generate a SQL schema for a blog"}]
)
print(response.choices[0].message.content)
Cost to start: $5, paid
Paid tier: GPT-4.1 mini at $0.40/$1.60 per 1M tokens, cached input $0.10
Best for: General-purpose tasks, function calling, structured outputs
That $0.40/$1.60 was the single number in the big August sweep that verified as still exactly correct. Everything else moved.
One housekeeping note: OpenAI's docs moved from platform.openai.com to developers.openai.com. Old links 301 to the new host, but if you've got docs URLs in a runbook, update them.
Together AI
This section used to say "$5 free credit." It doesn't anymore, because Together's own billing docs say: "Together AI does not currently offer free trials."
There is no credit. There's a $5 minimum purchase required before you can use the platform at all. Secondary sources still quote $5, $25, $50 and $100 credits, and the fact that they can't agree with each other is a decent tell that all of them are stale promos.
It's still a good service. Llama 3.3 70B serverless runs $1.04 per 1M tokens in both directions, and you get dozens of open models behind one API without touching a GPU. But it does not belong in a free-tier article as anything other than a correction, so I've pulled it from the free stack below.
Tier 3: Specialized Free Tiers
HuggingFace Inference
Restructured, and not in your favour. The old rate-limited-but-free serverless Inference API has been folded into "Inference Providers," and free accounts now get $0.10 per month in credit — a figure HuggingFace's own table marks "subject to change". PRO accounts get $2.00/month, Team and Enterprise get $2.00 per seat per month.
Ten cents a month is not a free tier for large models. It's a taste. This page previously showed a code sample hitting meta-llama/Llama-3.3-70B-Instruct through the old serverless endpoint, and I've removed it rather than leave a snippet that will burn your entire monthly allowance in a handful of calls. HuggingFace's own docs note that since July 2025 the hf-inference provider focuses mostly on CPU inference: embeddings, text ranking, text classification, smaller LLMs. Use it for what it's now scoped to, and read the pricing page before you wire anything up.
Replicate
Pay-per-prediction pricing. Correction (15 August): this page previously said Replicate offers "trial credits" — its pricing page mentions no credits of any kind, and what the billing docs actually say is "You can run select models on Replicate for free, but after a bit you'll be asked to set up billing." That's free runs of selected models, not a credit balance you draw down. Billing is per-second for hosted models and per-token for the LLMs it fronts. Claude on Replicate, for example, is $3.00 per million input tokens and $15.00 per million output tokens (Replicate lists the output price as $0.015 per thousand). That's more than the $2/$10 Anthropic charges directly for Claude Sonnet 5 — and the model behind Replicate's listing is anthropic/claude-3.7-sonnet, which Anthropic retired from its own first-party API on 19 February 2026, several generations behind what you'd get going direct.
Best for: Image generation, audio models, niche open-source models
Mistral
Whether the free "Experiment" tier still exists is itself no longer confirmable from Mistral's own pages: neither mistral.ai/pricing/api nor the models docs carries an Experiment-tier row, an RPM figure, or a tokens-per-day figure for it. The deployment tier page that used to carry those numbers now 404s. So the "1 RPM, 500K tokens/day" line this article ran in April is unconfirmable, and it's gone.
The current small model is specifically Mistral Small 4 (v26.03), a hybrid instruct/reasoning/coding model. Mistral's own docs point at Mistral Medium 3.5 as the flagship, not Small.
Cohere
Two corrections, both from the same table this page had garbled. The trial-key rate limit is 20 requests/minute on the Chat API. The volume cap is 1,000 API calls per month. Those are separate things and they belong in separate columns. The old "5 RPM / 100 calls per minute" pairing matched neither.
Other endpoints have their own trial limits: Embed at 2,000 inputs/min, Rerank at 10/min, Tokenize at 100/min, with a 500/min catch-all default. Cohere doesn't publish a token ceiling for trial keys separately from the 1,000-call cap.
Command R+ is no longer the flagship. Command A+ (command-a-plus-05-2026) is, per Cohere's models doc. Here's the odd part: Cohere's public pricing page doesn't list a price for Command A+ at all. Its price table stops at Command R+ 08-2024 at $2.50/$10.00 per 1M. So Cohere recommends a model it won't quote you a rate for, which usually means enterprise sales.
The Free Stack: Running AI for $0/Month
The honest version of this stack is shorter than it was in April.
| Layer | Provider | Why |
|---|---|---|
| Primary LLM | Gemini 2.5 Flash (free tier) | $0 in and out, check your console for limits |
| Fast inference | Groq (free tier) | 30 RPM, but watch the 1,000 req/day cap |
| Fallback LLM | Cohere trial key | 20 req/min, 1,000 calls/month |
| Embeddings | HuggingFace hf-inference
|
Now CPU-scoped, and only $0.10/month of credit |
| Image generation | Replicate (select models free) | Until you're asked to set up billing |
Together AI is out of this table this month for the reason above. HuggingFace stays only because embeddings are one of the few workloads that still fit inside ten cents, and even then I'd treat it as a stopgap. If you want more genuinely free infrastructure around this, free developer tool credits are still holding up better than the API tiers are.
Code Example: Smart Fallback Chain
async def generate_response(prompt: str) -> str:
providers = [
("gemini", call_gemini), # free tier, limits console-only
("groq", call_groq), # 30 RPM, 1,000 req/day
("cohere", call_cohere), # 20 req/min, 1,000 calls/month
]
for name, call_fn in providers:
try:
return await call_fn(prompt)
except RateLimitError:
logger.warning(f"{name} rate limited, falling back")
continue
raise Exception("All providers exhausted")
This pattern is how production AI apps work behind the scenes. Primary provider handles most requests. Fallbacks catch the rest. Log which provider served each request, because when a vendor quietly cuts a quota you want to see it in your own metrics before you read about it.
When to Pay
The free tiers hit their limits when you need:
- Sustained throughput above roughly 30 requests a minute
- The top models: Claude Opus 5, GPT-4.1, Gemini 3.1 Pro Preview
- Long-context processing across large codebases or long documents
- Production SLAs
For side projects and low-traffic apps, Gemini and Groq will genuinely carry you. For anything with real users, budget $50-200/month and stop counting requests.
One last thing, and it's the reason this page gets re-checked every week rather than quarterly. Between April and August, two of these ten vendors stopped publishing their free-tier numbers, one deleted its free trial, and one replaced a rate limit with a ten-cent budget. If you're reading a free-tier comparison that hasn't been touched in six months, assume most of it is fiction.
Related: Free Developer Tool Credits | Is the Claude API Free? | Best AI Tools Developers Actually Use
Top comments (0)