The APIs I created all show unavailable status, and when I try to request them, I get a 429 error. I don’t know what happened, but when I check Gemini API Usage, it shows No Data Available.
6 Likes
I have the same issue. It says the plan is “Unavailable“ instead of “Free“, despite me not having used it ever since the daily reset. And when I check into the console, it still says I have 0/50 messages used.
I’m also facing the same issue
Same here, I’m also getting the 429 error even though I haven’t hit the free token limit.
Any fix regarding this issue?
I decided to go to Google Cloud. It said something like, "Your API is blocked. Enable 2FA in your Google account.” I enabled 2FA, but my API still isn’t working.
1 Like
Try going to Google Cloud and see what is written there.
The best way to use Gemini API is through a billing enabled project/account. You don’t have to pay, plus Google offers free credits, and for your first 50k users you won’t even be charged. Billing, IAM rules and policies, and security rules are usually causing auth issues with the API. Might not be related to quote limits at all.
2 Likes
Unfortunately, I live in Russia and can’t do this. 
I feel for you. I am confident there are similar programs with Yandex Cloud where they can give you 5k-25k credits to build your first app, but I know for a fact their AI is nowhere close to Chinese or US models.
Try local models like ollama or check hugging face for models you can dl and are small, where the only restrictions would be your processing power. For text is usually nothing, unless you wanna generate multimedia content using gpu, or run it as a server for multiple users.
2 Likes
same here, i encountered it just today. all the api went to be unavailable
me too 
what happend ?
Same here. Creating new projects doesn’t help. No more free plan?
That’s unrelated to the API itself. The 2FA is to have access to your Google Cloud console. It doesn’t affect the projects and keys themselves. Some people are forced to enable 2FA and still can’t use their API, and some have it on and it still didn’t fix the issue. It seems like they changed something or added som additional action to take before being able to use your API but they didn’t bother telling us. It’s just like their rate limits page. It’s outdated and still says that the free pro is 100 messages per day, which is false, since they lowered it to 50 when you check in Google Cloud.
What are you making!!?
API??
It’s very simple.
Before you create an API, you must be able to determine its purpose and the measurement parameters. Don’t just create it, even if you’re using an iPhone 15. That doesn’t mean you can do everything. Hehe I’m sorry Ac is joking around
Understanding the Error Messages
A 429 Too Many Requests error is a real-time signal from the API’s server. It means that your client has sent more requests than the rate limit allows within a specific time frame. This error is instantaneous; the server actively rejects your request to prevent being overwhelmed.
On the other hand, the “No Data Available” message in the API Usage dashboard is not always real-time. Usage data is typically processed in batches, which can cause a reporting delay. The API is actively counting your requests and stopping you when you hit the limit, but that usage data takes time to show up in the dashboard.
Understanding the Error Messages
A 429 Too Many Requests error is a real-time signal from the API’s server. It means that your client has sent more requests than the rate limit allows within a specific time frame. This error is instantaneous; the server actively rejects your request to prevent being overwhelmed.
On the other hand, the “No Data Available” message in the API Usage dashboard is not always real-time. Usage data is typically processed in batches, which can cause a reporting delay. The API is actively counting your requests and stopping you when you hit the limit, but that usage data takes time to show up in the dashboard.
In short, you are hitting the limit, but the dashboard hasn’t caught up yet.
What to Do About It
Wait and Check Again: The most likely solution is to simply wait. Give it a few hours and then refresh the Gemini API Usage dashboard. The usage data should eventually populate, confirming that you did, in fact, exceed your quota.
Verify Your API Key: Double-check that the API key you are using for your requests belongs to the exact same Google Cloud project that you are viewing in the dashboard. Using a key from a different project will cause the dashboard to show zero usage for the project you’re on.
Review the Rate Limits: Look at the official documentation for the Gemini API. Find the specific rate limits for your tier (e.g., requests per minute, requests per day). This will help you understand exactly what the limits are so you can adjust your code to stay within them.
Implement Exponential Backoff: For your code, this is a standard practice for handling rate limits. If you receive a 429 error, wait a short period of time before retrying. If that retry also fails, wait for a longer period, and so on. This prevents you from continuously hitting the server with failed requests.
2 Likes
Bahasa Indonesia seperti ini
Memahami Pesan Kesalahan
Kesalahan 429 “Terlalu Banyak Permintaan” adalah sinyal waktu nyata dari server API. Ini berarti klien Anda telah mengirimkan lebih banyak permintaan daripada batas kuota yang diizinkan dalam jangka waktu tertentu. Kesalahan ini terjadi seketika; server secara aktif menolak permintaan Anda untuk mencegah kelebihan kapasitas.
Di sisi lain, pesan “Tidak Ada Data Tersedia” di dasbor Penggunaan API tidak selalu waktu nyata. Data penggunaan biasanya diproses secara batch, yang dapat menyebabkan keterlambatan pelaporan. API secara aktif menghitung permintaan Anda dan menghentikan Anda ketika Anda mencapai batas kuota, tetapi data penggunaan tersebut membutuhkan waktu untuk ditampilkan di dasbor.
Singkatnya, Anda memang mencapai batas kuota, tetapi dasbor belum merespons.
Apa yang Harus Dilakukan
Tunggu dan Periksa Kembali: Solusi yang paling mungkin adalah menunggu. Tunggu beberapa jam, lalu segarkan dasbor Penggunaan API Gemini. Data penggunaan pada akhirnya akan terisi, mengonfirmasi bahwa Anda memang telah melampaui kuota.
Verifikasi Kunci API Anda: Periksa kembali apakah kunci API yang Anda gunakan untuk permintaan Anda berasal dari proyek Google Cloud yang sama persis dengan yang Anda lihat di dasbor. Menggunakan kunci dari proyek lain akan menyebabkan dasbor menampilkan nol penggunaan untuk proyek yang sedang Anda ikuti.
Tinjau Batas Kecepatan: Lihat dokumentasi resmi untuk API Gemini. Temukan batas kecepatan spesifik untuk tingkatan Anda (misalnya, permintaan per menit, permintaan per hari). Ini akan membantu Anda memahami dengan tepat batasannya sehingga Anda dapat menyesuaikan kode Anda agar tetap berada di dalamnya.
Implementasi Eksponensial Backoff: Untuk kode Anda, ini adalah praktik standar untuk menangani batas kecepatan. Jika Anda menerima kesalahan 429, tunggu beberapa saat sebelum mencoba lagi. Jika percobaan ulang tersebut juga gagal, tunggu lebih lama, dan seterusnya. Ini mencegah Anda terus-menerus membebani server dengan permintaan yang gagal.
Agar penerapan berhasil,tidak perlu pengamanan, Acukan saja perihal pengamanan,insentisitas keamanan hanya berlaku pada jalur permulaan, dan yang menentukan keberhasilan finish adalah metode dan cara anda mengimplentasikanya. Seperti ini
Berikut adalah langkah-langkah untuk mengimplementasikannya:
-
Lakukan Permintaan Awal: Kirim permintaan API Anda seperti biasa.
-
Identifikasi Kode Kesalahan: Jika Anda menerima kesalahan yang menunjukkan server sedang sibuk atau kelebihan beban (misalnya kode
429atau503), jangan langsung coba lagi. Ini adalah tanda bahwa Anda perlu menunggu. -
Tentukan Waktu Tunggu Awal: Atur jeda waktu tunggu awal yang singkat, misalnya 1-2 detik.
-
Ulangi Permintaan dengan Jeda: Setelah jeda waktu awal berlalu, coba kembali mengirim permintaan.
-
Tingkatkan Jeda Secara Eksponensial: Jika permintaan kedua masih gagal, lipat gandakan waktu tunggu Anda. Misalnya, dari 1 detik menjadi 2, lalu 4, lalu 8, dan seterusnya. Peningkatan eksponensial ini memastikan Anda tidak terus-menerus mengirim permintaan dengan cepat ke server yang sedang kesulitan.
-
Tambahkan Jitter: Untuk membuat proses lebih acak dan mencegah semua klien mencoba kembali pada saat yang sama, tambahkan sedikit variasi acak (jitter) pada waktu tunggu Anda. Misalnya, alih-alih menunggu tepat 4 detik, Anda bisa menunggu antara 3.5 hingga 4.5 detik.
-
Tetapkan Batas Maksimum: Tentukan batas maksimal percobaan ulang atau total waktu tunggu. Ini akan mencegah aplikasi Anda terjebak dalam perulangan tak terbatas jika server benar-benar tidak tersedia.
we would like to pinpoint report this internally,
can you guyz please respond with following information,
Model name,
payment tier,
is your usage near the rate limits?
do you know which rate limit is being hit?(you can ususally find this in the 429 error report)
and any other info that can help us find the cause.
Thank you
gemini-2.5-pro
Free tier
**The usage is at 0.
The 429 error is broad. It only says resource exhausted. In the past, it specified error 429, you hit the limit per minute, but now, it just says in general error 429.
The odd thing is that when you create a new project, and create a new api key inside it, the first message always work, but then it hits the “error 429“**
Same problem - I get a new key and can’t send any messages. Problem since September 23.
Same problem with all gemini models
1 Like