Free AI for Open Source
A grant of 15,000,000 free AI tokens plus Pro-tier API access for maintainers of actively maintained open source projects.
100% Iimodeli ze-AI ezivulekileyo
Yonke imodeli ekhoyo ngokwayo kwi-Free.ai ivuliwe-ivela phantsi kwelayisensi ezivumelayo (Apache 2.0, MIT, BSD). Akukho Llama/Meta ilayisensi. Sikholelwa ukuba i-AI kufuneka ibonise, iqinisekiswe, kwaye ikhululekile ukuyisebenzisa.
Chat & Iimodeli Zolwimi
| Imodeli | Iinkqubo | Ilayisensi | Icebo |
|---|---|---|---|
| Qwen2.5-72B | Incoko yobuqu, umbono, uvavanyo | _Iifayile: | HuggingFace |
| Qwen2.5-7B | Name=Incoko yevidiyo ekhawulezayo, imisebenzi elula | _Iifayile: | HuggingFace |
| Mistral-7B | Incoko ekhawulezayo, injongo ebanzi | _Iifayile: | HuggingFace |
| Phi- 3 | Ukuqonda okusebenzayo, i-matematics | MIT | HuggingFace |
Iimodeli zekhowudi
| Imodeli | Iinkqubo | Ilayisensi | Icebo |
|---|---|---|---|
| Qwen2.5-Coder-32B | Uhlobo lwekhowudi, ukulungisa ingxaki | _Iifayile: | HuggingFace |
| DeepSeek-Coder-V2-Lite | Inkqubo yolawulo lweenkqubo | MIT | HuggingFace |
Imifanekiso & Ividiyo
| Imodeli | Iinkqubo | Ilayisensi | Icebo |
|---|---|---|---|
| FLUX.1-fast | Umfanekiso okhawulezayo | _Iifayile: | HuggingFace |
| Stable Diffusion XL | Umfanekiso ophezulu wobunjani | OpenRAIL++ | HuggingFace |
| _Iifayile: | Umfanekiso ovela kwi-Artist | _Iifayile: | HuggingFace |
| _Iifayile: | Ukwakha ividiyo kumbhalo | _Iifayile: | HuggingFace |
Speech & Audio Models
| Imodeli | Iinkqubo | Ilayisensi | Icebo |
|---|---|---|---|
| _Iifayile: | Umbhalo-usuka-kwi-lizwi | _Iifayile: | HuggingFace |
| _Iifayile: | Umbhalo-usuka-kwi-lizwi | MIT | GitHub |
| _Iifayile: | Umbhalo-usuka-kwi-lizwi | MIT | GitHub |
| _Iifayile: | Umbhalo-usuka-kwi-lizwi | MIT | HuggingFace |
| OutTTS | Umbhalo-usuka-kwi-lizwi | _Iifayile: | HuggingFace |
| _Iifayile: | Ukuthetha-ukuthetha-ukubhala | MIT | GitHub |
| I-Wax | Ukuthetha-ukuthetha-ukubhala | _Iifayile: | GitHub |
| ACE-Step 3.5B | Uhlobo loMculo | _Iifayile: | HuggingFace |
| Mel-Band RoFormer | Ukwahlula iingoma (uvocals, instrument) | MIT | arXiv |
| SCNet | Ukwahlula iingoma zengoma (i-drums, i-bass, ezinye) | _Iifayile: | arXiv |
Uguqulelo, Umbono & Ezinye Iimodeli
| Imodeli | Iinkqubo | Ilayisensi | Icebo |
|---|---|---|---|
| MadLAD- 400 3B | Uguqulelo kolunye ulwimi | _Iifayile: | HuggingFace |
| _Iifayile: | Umfanekiso onyukayo | BSD | GitHub |
| I-BRI RMBG 2. 0 | Isiqalo soSuku | _Iifayile: | HuggingFace |
| US | Umfanekiso | _Iifayile: | GitHub |
| Ukukhupha i-Oldify | Umbala wefoto | MIT | GitHub |
| all-MiniLM-L6-v2 | Iindawo ezingaphakathi koMbhalo | _Iifayile: | HuggingFace |
| I-Tesseract | OCR | _Iifayile: | GitHub |
Open Source Maintainer Program
Abalondolozi beprojekti ezigcinwe ngokusebenzayo ezivulekileyo ezivulekileyo banokufaka isicelo sonikezelo olukhululekileyo lwee-token ezili-15,000,000, unikezelo lwee-token zonyaka wePro, kunye ne-API key kwinqanaba lokufuna iPro (i-100,000 yesicelo ngenyanga). Unikezelo luquka yonke into esebenza kwimodeli yethu ekhoyo. Imodeli yangaphandle yepremiyamu ayinxalenye yonikezelo: isebenzisa i-token ethengiweyo.
Uluhlu lweeNkonzo
- Ugcina iprojekti (umnini okanye umnikezeli oqhelekileyo), kwaye indawo yokugcina ifumaneka ngokusemthethweni phantsi kwelayisensi yomthombo ovulekileyo eqinisekisiweyo ye-OSI
- Igcinwa ngokusebenzayo, ngee-commits okanye izikhumbuzo kwiintsuku ezili-90 ezidlulileyo
- Abanye abantu bayisebenzisa, ibonakaliswe nganye yazo: 50 iinkwenkwezi zeGitHub, 1,000 iipakeji ezilayishiweyo ngenyanga, okanye iiprojekthi ezixhomekeke kuyo
- One grant per Free.ai account across the open source, nonprofit and startup programs, added to an account you have already created
Ukufaka isicelo, thumela ikhonkco lovimba kunye nekhonkco lokubonisa ukusetyenziswa (iinkwenkwezi, ukukhuphela izibalo okanye abaxhomekeke) ukusuka kwidilesi ye-imeyile kwi-akhawunti yakho ye-Free.ai. Isicelo se-imeyili
Imibuzo ebuzwa rhoqo
Yes. Every model we run on our GPU servers is open-source under permissive licenses -- Apache 2.0, MIT, or BSD. We deliberately exclude models with restrictive licenses like Meta's Llama license. You can inspect, use, and modify all models freely.
Meta's Llama license includes restrictions on commercial use for large companies and requires specific attribution. We only use models with truly permissive licenses (Apache 2.0, MIT, BSD) that allow unrestricted commercial and personal use by anyone.
Every model in our table links to its source on HuggingFace or GitHub. You can download the weights directly from these repositories and run them on your own hardware.
Qwen2.5-72B is our most capable chat model for complex reasoning and analysis. Qwen2.5-7B and Mistral-7B are faster options for everyday tasks. Phi-3 is excellent for efficient reasoning and math.
We use five text-to-speech models: Kokoro (Apache 2.0), Piper (MIT), MeloTTS (MIT), Chatterbox (MIT), and OuteTTS (Apache 2.0). Each has different strengths in voice quality, speed, and language support.
FLUX.1-schnell for fast generation (Apache 2.0), Stable Diffusion XL for high-quality images (OpenRAIL++), and Kandinsky 2.2 for artistic styles (Apache 2.0). For video, we use CogVideoX-2B, Wan 2.2 and HunyuanVideo.
If you have developed or know of an open-source model under a permissive license that you think we should host, contact us. We evaluate models based on quality, license compatibility, and user demand.
I-free grant ye-15,000,000 tokens kunye ne-API keys kwinqanaba lesicelo sePro (100,000 isicelo ngenyanga) kubalondolozi beprojekti ezigcinwe ngokusebenzayo ezivulekileyo ezinobungqina bokusetyenziswa. Iquka izixhobo eziqhuba kwimodeli yethu ekhoyo. Imodeli yangaphandle yepremium ayinxalenye ye grant kwaye isebenzisa i-tokens ethengisiweyo. Faka ngekhonkco kwindawo yokugcina yakho kunye nobungqina bokusetyenziswa.
We evaluate every model before deploying it based on benchmark performance, real-world output quality, inference speed, and license compliance. Only models that meet our quality bar are added to the platform.
We use MadLAD-400 3B from Google, which supports 450+ languages under an Apache 2.0 license. It handles even low-resource languages that many commercial translation services do not support.
Real-ESRGAN (BSD) for image upscaling, BRIA RMBG 2.0 (Apache 2.0) for background removal, SAM (Apache 2.0) for image segmentation, and DeOldify (MIT) for photo colorization.
Hayi: iimodyuli zangaphandle eziphezulu ezinjenge GPT-5, Claude ne Gemini ziyipropathi, zithwala iimeko zabo zokufaka ilayisensi, ziqhuba kwi-token ethengiweyo kwaye aziyinxalenye ye-comp grant. Iimodyuli ezihonjiswayo kuphela ezidweliswe kule phepha zi open source.