---
title: "Requesty Free Models"
slug: "requesty-free-models"
provider: "Requesty"
status: "verified"
verifiedAt: "2026-10-12"
listedAt: "2026-09-12"
canonical: "https://aifree.dev/offer/requesty-free-models"
---

# Requesty Free Models

> 13 free LLMs at $0 with 200 requests a day via Requesty's gateway

- Provider: [Requesty](https://aifree.dev/provider/requesty-ai)
- What is free: 200 requests/day
- Status: verified by a person on 2026-10-12
- Credit card: not required
- Sign-up: required
- For: Developers
- Resets: daily
- Claim at: https://www.requesty.ai/models?free=1

## What you get

Requesty's LLM gateway serves 13 models free at $0 in/out pricing (Nemotron, Gemma, Laguna, and more) with 200 requests/day on the free plan — OpenAI-compatible API, no credit card required.

## The catch

Only the 13 designated zero-cost models are free; they are rate-limited per organization and free-model terms can change.

## Limits

- Free plan: 200 requests/day, no credit card required
- 13 models at $0 in/out pricing
- Free models: nvidia/nemotron-3-ultra-550b-a55b, nvidia/nemotron-3-super-120b-a12b, nvidia/nemotron-3.5-lightning-30b-a3b, nvidia/nemotron-3-nano-omni-30b-a3b-reasoning, nvidia/nemotron-3-nano-30b-a3b, nvidia/nemotron-3.5-content-safety, meta/muse-glimmer-30b, inclusionai/ling-3.0-tiny, poolside/laguna-m.1, poolside/laguna-xs.2, mistral/leanstral-1-5, google/gemma-4-31b-it, inclusionai/ling-3.1-flash

## Models served free

Published free-tier rate limits; "—" = not published.

| Model | API id | Context (k tokens) | Tool calling | Vision | RPM | RPD | TPM |
| --- | --- | --- | --- | --- | --- | --- | --- |
| nvidia/nemotron-3-ultra-550b-a55b | — | 1000 | yes | — | 20 | 200 | — |
| nvidia/nemotron-3.5-lightning-30b-a3b | — | 1000 | yes | — | 20 | 200 | — |
| google/gemma-4-31b-it | — | 262 | yes | yes | 20 | 200 | — |
| inclusionai/ling-3.1-flash | — | 262 | yes | — | 20 | 200 | — |
| mistral/leanstral-1-5 | — | 262 | yes | — | 20 | 200 | — |
| meta/muse-glimmer-30b | — | 131 | yes | yes | 20 | 200 | — |

## Free forever by self-hosting

[Ollama](https://github.com/ollama/ollama) (MIT)

## Details

## Overview

[Requesty](https://www.requesty.ai/models?free=1) is an LLM gateway with an OpenAI-compatible API (swap the base URL to `https://router.requesty.ai/v1` — no SDK changes). It routes a catalog of 600+ models from 30 labs, of which 13 are currently free to call at $0 input and output pricing. Free models run through the same gateway as paid ones, so routing, caching, fallbacks, logging, and analytics all apply.

## Free Tier Quotas & Limits

- **Free plan:** access to all free models, 200 requests per day, no credit card required. Free models are rate-limited per organization: new organizations get 200 requests/day and 20 requests/minute; paying organizations get 1,000 requests/day and 60 requests/minute. Limits apply across all free models combined and reset daily; hitting them returns a rate-limit error on free models only.
- New accounts also include free credits to start routing immediately (amount not specified).
- Pay-as-you-go pricing adds a 5% markup over provider rates (0% when bringing your own keys).
- The docs note these models are "free for now" — pricing changes will be announced in the changelog ahead of time.
- Cached input tokens are not free on every model (e.g. `muse-glimmer-30b` lists $0.04/M cached, `nemotron-3-ultra-550b-a55b` lists $1.00/M cached).

## Key Free Models

Gateway identifiers (per the live catalog; the docs page also lists all 13):

- `nvidia/nemotron-3-ultra-550b-a55b` (1.0M context)
- `nvidia/nemotron-3-super-120b-a12b` (1.0M)
- `nvidia/nemotron-3.5-lightning-30b-a3b` (1.0M)
- `nvidia/nemotron-3-nano-omni-30b-a3b-reasoning` (131K)
- `nvidia/nemotron-3-nano-30b-a3b` (262K)
- `nvidia/nemotron-3.5-content-safety` (131K)
- `meta/muse-glimmer-30b` (131K)
- `inclusionai/ling-3.0-tiny` (262K)
- `poolside/laguna-m.1` (33K)
- `poolside/laguna-xs.2` (33K)
- `mistral/leanstral-1-5` (262K)
- `google/gemma-4-31b-it` (262K)
- `inclusionai/ling-3.1-flash` (262K)

## How to Get Started

1. Sign up at [app.requesty.ai](https://app.requesty.ai) and create an API key.
2. Point any OpenAI-compatible SDK at `https://router.requesty.ai/v1` (the Anthropic SDK works against `/anthropic/v1/messages`).
3. Call a free model by ID, e.g. `model="nvidia/nemotron-3-super-120b-a12b"`.

## More

- HTML page: https://aifree.dev/offer/requesty-free-models
- Change history: https://aifree.dev/offer/requesty-free-models/history
- How it was verified: https://aifree.dev/offer/requesty-free-models/verification
- More in [Free AI API credits](https://aifree.dev/free-api-credits.md)
- JSON: https://aifree.dev/api/v1/offers/requesty-free-models