Compare free API access by the work it covers and the conditions that end it. Keep a prototype’s model identity, eligibility and stopping behavior explicit.
What matters for free ai apis
Distinguish credit grants, model-specific quota and free model routes. Each can support an experiment, but they describe different account conditions. Read the current source and confirm the offer in the account that will make the call before relying on it.
Choose a small safe fixture with a verifiable answer. A free experiment is most useful when every call answers a question about integration or quality. Include a missing-evidence case and inspect whether the model can stop or communicate uncertainty instead of inventing a result.
OpenRouter’s free router selects eligible free candidates. Official documentation. Model Studio’s offer uses model-specific quotas with defined scope. Official documentation.
A free-only product needs a failure state when access is unavailable. Decide whether work should pause, be queued within a deadline or stop. A paid fallback belongs in configuration only when that continuation is intentional and budgeted.
Ranked candidates
Only active candidates with documented values for this ranking are included. Prices retain the tier and deployment condition shown below.
| Model | Provider | Input USD / 1M | Output USD / 1M | Context tokens | Price condition | Cost |
|---|---|---|---|---|---|---|
| Cohere: North Mini Code (free) | OpenRouter | Free | Free | 256,000 | Standard | Estimate cost |
| Dots Studio: Dots3-Note Preview (free) | OpenRouter | Free | Free | 512,000 | Standard | Estimate cost |
| Free Models Router | OpenRouter | Free | Free | 200,000 | Standard | Estimate cost |
| Gemini 2.5 Flash | Free | Free | 1,048,576 | free tier | Estimate cost | |
| Gemini 2.5 Flash Preview TTS | Free | Free | 8,192 | free tier | Estimate cost | |
| Gemini 2.5 Flash-Lite | Free | Free | 1,048,576 | free tier | Estimate cost | |
| Gemini 2.5 Pro | Free | Free | 1,048,576 | free tier | Estimate cost | |
| Gemini 3 Flash Preview | Free | Free | 1,048,576 | free tier | Estimate cost | |
| Gemini 3.1 Flash TTS Preview | Free | Free | 8,192 | free tier | Estimate cost | |
| Gemini 3.1 Flash-Lite | Free | Free | 1,048,576 | free tier | Estimate cost | |
| Gemini 3.5 Flash | Free | Free | 1,048,576 | free tier | Estimate cost | |
| Gemini 3.5 Flash-Lite | Free | Free | 1,048,576 | free tier | Estimate cost | |
| Gemini 3.5 Live Translate | Free | Free | 16,384 | free tier | Estimate cost | |
| Gemini 3.5 Transcribe | Free | Free | 98,304 | free tier | Estimate cost | |
| Gemini 3.5 Transcribe Live | Free | Free | 131,072 | free tier | Estimate cost | |
| Gemini 3.6 Flash | Free | Free | 1,048,576 | free tier | Estimate cost | |
| Gemini 3.7 Flash | Free | Free | 1,048,576 | free tier | Estimate cost | |
| Gemini 3.8 Flash | Free | Free | 1,048,576 | free tier | Estimate cost | |
| Google: Gemma 4 26B A4B (free) | OpenRouter | Free | Free | 262,144 | Standard | Estimate cost |
| Google: Gemma 4 31B (free) | OpenRouter | Free | Free | 262,144 | Standard | Estimate cost |
Last verified · Source ↗
| Condition | Verified details |
|---|---|
| Current terms | The official model catalog lists models with zero input and output token rates. Availability and free-model request limits can change. |
| Eligible models | inclusionai/ling-3.0-flash-vl:free, nex-agi/nex-n2.5-mini:free, nex-agi/nex-n2.5-pro:free, inclusionai/ling-3.0-flash-sante:free, inclusionai/ling-3.0-flash-fin:free, dots-studio/dots-3-note-preview:free, liquid/lfm-2.5-2.6b:free, nvidia/nemotron-3.5-lightning:free, thinkingmachines/inkling-small:free, poolside/laguna-s-2.1:free, thinkingmachines/inkling:free, poolside/laguna-xs-2.1:free, cohere/north-mini-code:free, nvidia/nemotron-3.5-content-safety:free, nvidia/nemotron-3-ultra-550b-a55b:free, nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free, google/gemma-4-26b-a4b-it:free, google/gemma-4-31b-it:free, google/lyria-3-pro-preview, google/lyria-3-clip-preview, nvidia/nemotron-3-super-120b-a12b:free, openrouter/free |
Last verified · Source ↗
| Condition | Verified details |
|---|---|
| Current terms | Selected models have a Free Tier in the official pricing table. Paid-only features and use of submitted data follow the terms linked on that page. |
| Eligible models | gemini-3.8-flash, gemini-3.7-flash, gemini-3.6-flash, gemini-3.5-flash, gemini-3.5-live-translate-preview, gemini-3.5-transcribe-live, gemini-3.5-transcribe, gemini-3.5-flash-lite, gemini-3.1-flash-lite, gemini-3.1-flash-tts-preview, gemini-3-flash-preview, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, gemini-2.5-flash-preview-tts, gemini-robotics-er-2, gemini-robotics-er-2-streaming |
Last verified · Source ↗
| Condition | Verified details |
|---|---|
| Current terms | The official rate-limit reference publishes Free Plan limits. The available models and quotas are shown in that table; entitlement is checked in your account. |
Last verified · Source ↗
| Condition | Verified details |
|---|---|
| Current terms | Eligible newly activated Model Studio accounts can receive model-specific token quotas. Eligibility, model coverage, and expiration are defined by the official quota page. |
| Expiry | 90 days |
Last verified · Source ↗
The official billing documentation was checked. No universal ongoing free inference allowance was established by that source. Promotional or account-specific credits require a check in your account. Read the official terms ↗
Last verified · Source ↗
The official billing documentation was checked. No universal ongoing free inference allowance was established by that source. Promotional or account-specific credits require a check in your account. Read the official terms ↗
Last verified · Source ↗
The official billing documentation was checked. No universal ongoing free inference allowance was established by that source. Promotional or account-specific credits require a check in your account. Read the official terms ↗
Last verified · Source ↗
The official billing documentation was checked. No universal ongoing free inference allowance was established by that source. Promotional or account-specific credits require a check in your account. Read the official terms ↗
Last verified · Source ↗
The official billing documentation was checked. No universal ongoing free inference allowance was established by that source. Promotional or account-specific credits require a check in your account. Read the official terms ↗
Last verified · Source ↗
The official billing documentation was checked. No universal ongoing free inference allowance was established by that source. Promotional or account-specific credits require a check in your account. Read the official terms ↗
Last verified · Source ↗
The table identifies documented zero-priced model categories; the offer blocks describe account entitlements separately. Do not infer an ongoing free tier from a missing price row or a provider that has not published an applicable offer.
Our three picks
Find your starting point
Your text and estimates stay in this browser. No API requests are sent to model providers.
Loading verified model records…
The cheapest role can identify a documented free path that matches the requested capabilities. Balanced and strongest remain evaluation roles: an offer’s size or a model’s context capacity does not prove output quality. Check the actual candidate and its serving policy before introducing application data.
When using a free router, save the model that actually served each response. Evaluate results by candidate rather than pooling changing models into a supposedly repeatable benchmark. If a fixed model matters, choose an available documented variant and verify its current status.
When each pick is wrong
A free candidate is wrong when its terms, input support or availability do not fit the task. A free route can be suitable for learning while failing a production deadline. Do not compensate by silently sending the same workload to a paid route.
An expiring offer is wrong as the sole financial basis for continuous service. Keep a paid scenario or an intentional shutdown plan, and verify what happens to queued work at the boundary. Stop abandoned jobs when the experiment ends.
Read free offers explained and compare Model Studio quota scope with OpenRouter free routes.
Estimate cost
Estimate your API costs
Your text and estimates stay in this browser. No API requests are sent to model providers.
Loading verified model records…
Use the calculator to preserve the paid continuation scenario, including output and additional model stages.
Retain observed usage even while an offer covers the charge. It is the evidence needed to decide whether the application remains affordable when the offer changes or expires.
Last verified · Source ↗
Frequently asked questions
Is every free offer an unlimited free API?
What is the difference between a free route and credit?
Can a free router change the model?
Should a free-only flow use paid fallback automatically?
Why estimate paid usage during a free experiment?
What should happen at experiment closeout?
Sources
Last verified · Source ↗