Which cards can I pay with?

Visa, Mastercard and Meeza cards, and Apple Pay. Every payment is charged in Egyptian pounds.

Which mobile wallets can I pay with?

Vodafone Cash, Etisalat Cash, Orange Cash and WE Pay.

How fast does my credit arrive?

Credit is added to your balance once Kashier confirms the payment as successful, normally within a few seconds of checkout.

Pricing

AI API pricing, billed in EGP

Every model on one API, OpenAI-compatible for chat, embeddings and images, priced in Egyptian pounds and paid from a prepaid EGP balance, so you never pay in dollars. Text models are billed per token used, image models per generated image. Rates for models served by international providers are set from their USD prices at the current exchange rate, so they can change; this page shows the latest rates.

235 live models · input from 0.10 to 1765.57 EGP per million tokens

Building something? Developer Packs include free tokens every day on our sovereign models.

Studying at a university? The Student plan gives you free use every day on our sovereign models.

Developer Packs

A set price for a set number of days, with free tokens every day on our sovereign models. Other models, and by default use past the daily tokens, are paid from your balance at the per-token rates below.

Developer Pack

You pay

129.00 EGP for 30 days

4.30 EGP a day

You get every day

About 5.6M free tokens on Qwen3.8-27B-FP8, worth 70.00 EGP at standard prices.

Resets at 00:00 Cairo time; unused tokens don't carry over.

120 requests a minute · 4 at a time · replies up to 32,000 tokens · prompts up to 100,000 tokens

Get this pack

You buy a pack from your prepaid balance, on the Plans page of the dashboard. Any credit counts, welcome credit and coupons included.
Auto-renew is off unless you turn it on when you buy. You can turn it off at any time; the pack then runs to its end date. Unused days are not refunded.
At busy times, requests on the free daily amount can be asked to retry later.
Token counts are estimates: they assume 3 input tokens for every output token. Output tokens cost more, so long replies use the daily tokens faster.

Student plan

Free for university students. About 1.9M tokens a day on Qwen3.8-27B-FP8, for 12 months.

The free amount renews at 00:00 Cairo time; what's unused doesn't carry over.

20 requests a minute · 2 at a time · replies up to 16,384 tokens · prompts up to 100,000 tokens

Supported universities: Every Egyptian university (any address ending in .edu.eg), An-Najah National University, Nablus, Palestine (stu.najah.edu), Arab Academy for Science, Technology and Maritime Transport (aast.edu), The American University in Cairo (aucegypt.edu).

Don't see your university? Tell us at support@sovereigneg.com.

Sign up, then open Billing → Plans in your dashboard: verify your phone number there, then your university email: in one click if you signed up with it, otherwise with a 6-digit code we send to it. Personal accounts only.
A university email can verify one account only, and stays linked to it even if that account is deleted.
To keep it, verify your university email again before it ends: we email you a reminder.
Other models, and by default use past the daily amount, are paid from your balance at the per-token rates below.
At busy times, requests on the free daily amount can be asked to retry later.

Get started Already have an account? Verify on the Plans page

Chat models

177 models

Billed per million tokens, input and output priced separately. Rates shown are for each model's primary provider: a request served by another provider after a failover is billed at that provider's rates, which can be higher or lower, and cached, audio and image tokens can carry their own rates. Your request logs show what each request cost.

Model	Context	Input EGP / 1M	Output EGP / 1M
meta-llama-3.1-8b-instruct-turboMeta	131K	1.18	2.35
nim/meta/llama-3.1-8b-instructnim	131K	1.18	2.35
Meta Llama 3.2 1B InstructMeta	60K	1.59	11.83
mistral-nemo-instruct-2407Mistral AI	131K	1.71	1.77
gpt-oss-20bOpenAI	131K	1.77	8.24
Qwen: Qwen3.7 FlashAlibaba	1M	1.77	7.65
gpt-oss-120bOpenAI	131K	2.18	10.00
gemma-3-12b-itGoogle	131K	2.94	8.83
gemma-3-4b-itGoogle	131K	2.94	5.89
gpt-5-nanoOpenAI	400K	2.94	23.54
Meta Llama 3.2 3B InstructMeta	131K	2.94	19.42
mistral-small-24b-instruct-2501Mistral AI	33K	2.94	4.71
DeepSeek: DeepSeek V4 Flash 0731DeepSeek	1M	3.53	10.59
granite-4.2-8bibm-granite	131K	3.53	14.71
Ling-3.0-flashinclusionai	131K	3.53	10.59
ling-3.0-flash-vlinclusionAI	131K	3.53	10.59
nemotron-3-nano-30b-a3bNVIDIA	262K	3.53	14.12
nvidia-nemotron-3.5-lightningNVIDIA	262K	3.53	9.42
glm-4.7-flashzai-org	131K	3.56	23.54
Qwen: Qwen3.5-FlashAlibaba	1M	3.83	15.30
gemma-4-26b-a4b-itGoogle	262K	4.12	20.01
phi-4Microsoft	16K	4.12	8.24
mistral-small-3.2-24b-instruct-2506Mistral AI	128K	4.41	11.77
nvidia-nemotron-3-super-120b-a12bNVIDIA	262K	4.41	23.54
gemma-3-27b-itGoogle	131K	4.71	9.42
qwen3-32bAlibaba	41K	4.71	16.48
Qwen3-0.6BAlibaba	4K	5.00	10.00
deepseek-v4-flashDeepSeek	1M	5.30	10.59
gemma-4-31b-it-turboGoogle	262K	5.30	20.01
qwen3-235b-a22b-instruct-2507Alibaba	262K	5.30	32.37
qwen3-next-80b-a3b-instructAlibaba	262K	5.30	64.74
deepseek-v4.1-flash-1017DeepSeek	1M	5.77	17.30
Arize AI Qwen 2 1.5B Instructarize-ai	33K	5.89	5.89
Claude Haiku 5.5Anthropic	1M	5.89	29.43
gpt-6-lunaOpenAI	1M	5.89	29.43
llama-3.3-70b-instruct-turboMeta	131K	5.89	18.83
llama-4-scout-17b-16e-instructMeta	328K	5.89	17.66
qwen3.5-9bAlibaba	262K	5.89	8.83
qwen3.6-35b-a3bAlibaba	262K	5.89	55.91
seed-2.0-miniByteDance	256K	5.89	23.54
voxtral-small-24b-2507Mistral AI	33K	5.89	17.66
Qwen: Qwen3.8 FlashAlibaba	1M	6.65	22.48
Qwen: Qwen3 Coder NextAlibaba	262K	7.06	47.08
qwen3-14bAlibaba	41K	7.06	14.12
qwen3-30b-a3bAlibaba	41K	7.06	29.43
hy3tencent	262K	7.77	31.07
Qwen3.8-27B-FP8Alibaba	131K	8.21	25.00
deepseek-v4.1-flashDeepSeek	1M	8.24	24.72
qwen3.5-35b-a3bAlibaba	262K	8.24	58.85
Xiaomi: MiMo-V2.6-Flashxiaomi	1M	8.24	16.48
glm-5.3-flashzai-org	1M	8.83	29.43
gpt-4o-miniOpenAI	128K	8.83	35.31
gpt-4o-mini-2024-07-18OpenAI	128K	8.83	35.31
MiniMax: MiniMax M3.1 Flashminimax	1M	8.83	35.31
NVIDIA Nemotron 3.5 Super VLNVIDIA	262K	8.83	29.43
Qwen: Qwen3.8 27BAlibaba	262K	8.83	110.35
Qwen3 Coder 30B A3b InstructAlibaba	262K	8.83	35.31
qwen3-vl-30b-a3b-instructAlibaba	262K	8.83	35.31
llama-guard-4-12bMeta	164K	10.59	10.59
Qwen: Qwen3.6 FlashAlibaba	1M	11.03	66.21
Qwen: Qwen3 Coder FlashAlibaba	1M	11.48	57.38
gemma-4-31b-itGoogle	262K	11.77	23.54
gpt-5.4-nanoOpenAI	400K	11.77	73.57
gpt-5.6-lunaOpenAI	1M	11.77	70.62
nemotron-content-safety-3.5NVIDIA	131K	11.77	11.77
qwen3-vl-235b-a22b-instructAlibaba	262K	11.77	51.79
Ripple 65B Previewricursive-ai	131K	11.77	35.31
step-3.7-flashstepfun-ai	256K	11.77	67.68
minimax-m2.7MiniMaxAI	197K	12.36	49.44
DeepSeek: DeepSeek V4 Flash Vision ExpDeepSeek	1M	12.69	38.07
deepseek-v3.1DeepSeek	164K	14.71	55.91
DeepSeek: DeepSeek V3.1DeepSeek	164K	14.71	55.91
gemini-3.1-flash-liteGoogle	1M	14.71	88.28
gpt-5-miniOpenAI	400K	14.71	117.70
Qwen3-VL-8B-InstructAlibaba	262K	14.71	44.14
seed-1.8ByteDance	256K	14.71	117.70
deepseek-v3.2DeepSeek	164K	15.30	22.36
Qwen: Qwen3.5 Plus 2026-02-15Alibaba	1M	15.30	91.81
qwen3.5-122b-a10bAlibaba	262K	15.30	122.41
qwen3.5-27bAlibaba	262K	15.30	153.02
deepseek-v3.1-terminusDeepSeek	164K	15.89	58.85
gemma-4-31b-it-ultraGoogle	131K	15.89	44.73
minimax-m3MiniMaxAI	524K	16.48	64.74
DeepSeek: DeepSeek V3 0324DeepSeek	128K	17.07	67.09
gemini-2.5-flashGoogle	1M	17.66	147.13
Inkling 2 Smallthinkingmachines	1M	17.66	58.85
Qwen: Qwen3.5 Plus 2026-04-20Alibaba	1M	17.66	105.93
deepseek-v3DeepSeek	164K	18.83	52.38
qwen3.6-27bAlibaba	262K	18.83	188.33
Qwen3.7 PlusAlibaba	1M	18.83	75.33
Muse Glimmer 30Bmeta-models	131K	20.60	88.28
Z.ai: GLM 5.3 FlashXz-ai	1M	21.78	73.57
Xiaomi: MiMo-V2.6-Proxiaomi	1M	22.78	46.08
Beam 1.0 501B A23Breflection	1M	23.54	70.62
glm-4.7zai-org	203K	23.54	102.99
gpt-4.1-miniOpenAI	1M	23.54	94.16
mimo-v2.5XiaomiMiMo	1M	23.54	117.70
mimo-v2.5-proXiaomiMiMo	1M	25.60	51.20
qwen3-235b-a22b-thinking-2507Alibaba	128K	26.48	205.98
qwen3.5-397b-a17bAlibaba	262K	26.48	176.56
Thinking Machines: Inkling Smallthinkingmachines	524K	26.48	70.62
deepseek-r1-0528DeepSeek	164K	29.43	126.53
glm-4.6zai-org	203K	29.43	117.70
kimi-k2.5moonshotai	262K	29.43	147.13
nvidia-nemotron-3-ultra-550b-a55bNVIDIA	262K	29.43	129.47
seed-2.0-codeByteDance	256K	29.43	176.56
seed-2.0-proByteDance	256K	29.43	176.56
glm-5.2zai-org	1M	33.10	105.93
Z.ai: GLM 5.3z-ai	1M	33.10	147.13
MoonshotAI: Kimi K2 0711moonshotai	131K	33.55	135.36
Elorian Glimpse 1.0elorian-labs	262K	35.31	176.56
GLM 4.5Vzai-org	66K	35.31	105.93
kimi-k2.7-codemoonshotai	262K	35.31	205.98
MoonshotAI: Kimi K2 0905moonshotai	262K	35.31	147.13
NVIDIA Nemotron 3 Ultra 550B A55B NVFP4NVIDIA	203K	35.31	141.25
Gemma-2 Instruct (27B)Google	8K	38.25	38.25
nim/meta/llama-3.3-70b-instructnim	128K	41.79	41.79
nim/meta/llama-3.1-70b-instructnim	131K	42.37	42.37
Google: Gemini 3.6 FlashGoogle	1M	44.14	220.70
Google: Gemini 3.7 FlashGoogle	1M	44.14	220.70
Google: Gemini 3.8 FlashGoogle	1M	44.14	220.70
gpt-5.4-miniOpenAI	400K	44.14	264.84
kimi-k2.6moonshotai	262K	44.14	205.98
Qwen2.5-VL (72B) InstructAlibaba	128K	47.08	58.85
Tencent: Hy4 previewtencent	1M	49.08	147.19
claude-haiku-4.5Anthropic	200K	58.85	294.26
glm-5zai-org	203K	58.85	188.33
Inkling FP4thinkingmachines	524K	58.85	238.35
OpenAI: GPT-6.1 TerraOpenAI	1M	58.85	294.26
gemini-2.5-proGoogle	1M	73.57	588.52
gpt-5OpenAI	400K	73.57	588.52
gpt-5.1OpenAI	400K	73.57	588.52
Muse Spark Open 816B A42Bmeta-models	1M	73.57	250.12
deepseek-v4-proDeepSeek	1M	76.51	153.02
DeepSeek: DeepSeek V4 Pro 0813DeepSeek	1M	76.51	153.02
glm-5.1zai-org	203K	82.39	258.95
gemini-3.5-flashGoogle	1M	88.28	529.67
Qwen: Qwen3.8 MaxAlibaba	256K	97.11	291.38
gpt-5.2OpenAI	400K	102.99	823.93
MoonshotAI: Kimi K3.1 Promoonshotai	1M	111.82	559.10
Claude Sonnet 5.5Anthropic	1M	117.70	588.52
claude-sonnet-5Anthropic	1M	117.70	588.52
gemini-3.1-proGoogle	1M	117.70	706.23
Google: Gemini 4 ArgonGoogle	1M	117.70	588.52
Google: Gemini 4 CarbonGoogle	1M	117.70	588.52
gpt-4.1OpenAI	1M	117.70	470.82
gpt-5.6-terraOpenAI	1M	117.70	706.23
gpt-6-solOpenAI	1M	117.70	588.52
gpt-6.1-solOpenAI	1M	117.70	588.52
o3OpenAI	200K	117.70	470.82
qwen3.8-2.4t-a95bAlibaba	1M	117.70	353.11
SpaceXAI: Grok 4.6x-ai	500K	117.70	353.11
SpaceXAI: Grok 4.7x-ai	500K	117.70	353.11
xAI: Grok 4.5x-ai	500K	117.70	353.11
gpt-4oOpenAI	128K	147.13	588.52
gpt-4o-2024-08-06OpenAI	128K	147.13	588.52
gpt-4o-2024-11-20OpenAI	128K	147.13	588.52
gpt-5.4OpenAI	1M	147.13	882.78
MoonshotAI: Kimi K3moonshotai	1M	167.73	838.64
Claude Sonnet 4.5Anthropic	200K	176.56	882.78
claude-sonnet-4.6Anthropic	1M	176.56	882.78
gpt-3.5-turbo-16kOpenAI	16K	176.56	235.41
Claude Opus 5.5Anthropic	1M	235.41	1177.05
gpt-5.6-solOpenAI	1M	235.41	1177.05
Meta: Muse WonderMeta	1M	235.41	1177.05
Claude Opus 4.5Anthropic	200K	294.26	1471.31
Claude Opus 4.6Anthropic	1M	294.26	1471.31
Claude Opus 5Anthropic	1M	294.26	1471.31
claude-opus-4.7Anthropic	1M	294.26	1471.31
claude-opus-4.8Anthropic	1M	294.26	1471.31
gpt-4o-2024-05-13OpenAI	128K	294.26	882.78
gpt-5.5OpenAI	1M	294.26	1765.57
Claude Fable 5.2Anthropic	1M	470.82	2354.09
Claude Fable 5.1Anthropic	1M	588.52	2942.61
claude-fable-5Anthropic	1M	588.52	2942.61
gpt-6-astraOpenAI	1M	588.52	2942.61
gpt-4OpenAI	8K	1765.57	3531.14

Embedding models

29 models

Billed on input tokens only. Embeddings return vectors, not generated text, so there is no output charge.

Model	Context	Input EGP / 1M
embeddinggemma-300mSovereignEG	—	0.10
embeddinggemma-300mGoogle	2K	0.12
all-minilm-l12-v2SBERT	512	0.29
all-minilm-l6-v2SBERT	512	0.29
all-mpnet-base-v2SBERT	512	0.29
bge-base-en-v1.5BAAI	512	0.29
clip-vit-b-32SBERT	77	0.29
clip-vit-b-32-multilingual-v1SBERT	512	0.29
e5-base-v2intfloat	512	0.29
gte-basethenlper	512	0.29
multi-qa-mpnet-base-dot-v1SBERT	512	0.29
paraphrase-minilm-l6-v2SBERT	512	0.29
text2vec-base-chineseshibing624	512	0.29
bge-en-iclBAAI	8K	0.59
bge-large-en-v1.5BAAI	512	0.59
bge-m3BAAI	8K	0.59
bge-m3-multiBAAI	512	0.59
bge-m3-multi-8kBAAI	8K	0.59
e5-large-v2intfloat	512	0.59
gte-largethenlper	512	0.59
llama-nemotron-embed-vl-1b-v2NVIDIA	10K	0.59
multilingual-e5-largeintfloat	512	0.59
multilingual-e5-large-instructintfloat	512	0.59
qwen3-embedding-0.6bAlibaba	33K	0.59
qwen3-embedding-8bAlibaba	33K	0.59
qwen3-embedding-4bAlibaba	33K	1.18
text-embedding-3-smallOpenAI	—	1.18
text-embedding-ada-002OpenAI	—	5.89
text-embedding-3-largeOpenAI	—	7.65

Rerank models

4 models

Billed once per request on /v1/rerank: per 1M input tokens (the query plus the documents), or per search, where one search is one query with up to 100 documents and a document longer than 500 tokens together with the query counts as more than one. There is no output charge. Every response reports the EGP charge in usage.cost.

Model	Context	Price
Qwen3 Reranker 8BAlibaba	41K	11.77 EGP per 1M input tokens
Cohere: Rerank 4 Fastcohere	33K	0.1284 EGP per search
Cohere: Rerank 4 Procohere	33K	0.1605 EGP per search
Cohere: Rerank v3.5cohere	4K	0.0642 EGP per search

Image generation models

22 models

Billed for each image returned. A figure that names a size (for example 'per 1024p image') is the price at that size, and the charge scales with the pixels you request; other figures are per image. ≈ marks the provider's published default price, so the exact charge depends on the request. Every response reports the EGP charge in usage.cost.

Model	Price
sdxl-turbostabilityai	0.0118 EGP per 1024p image
flux-1-schnellblack-forest-labs	0.0294 EGP per 1024p image
p-imagePrunaAI	0.2943 EGP per image
flux-1-devblack-forest-labs	0.5297 EGP per 1024p image
flux-2-klein-4bblack-forest-labs	0.8239 EGP per 1024p image
flux-2-klein-9bblack-forest-labs	0.8828 EGP per 1024p image
flux-2-problack-forest-labs	≈ 0.8828 EGP per image
flux-2-devblack-forest-labs	1.05 EGP per 1024p image
wan2.6-t2iWan-AI	1.77 EGP per image
nano-banana-2-liteGoogle	1.98 EGP per image
bria-3.2Bria	2.35 EGP per image
bria-3.2-vectorBria	2.35 EGP per image
fibo-2.0Bria	2.35 EGP per image
flux-1.1-problack-forest-labs	2.35 EGP per image
seedream-4ByteDance	2.35 EGP per image
seedream-4.5ByteDance	2.35 EGP per image
nano-banana-2Google	3.95 EGP per image
flux-2-maxblack-forest-labs	≈ 4.12 EGP per image
qwen-image-maxAlibaba	4.41 EGP per image
seedream-5.0-proByteDance	5.83 EGP per image
gemini-3-pro-imageGoogle	7.91 EGP per image
nano-banana-proGoogle	7.91 EGP per image

Decision models

3 models

System One models on /v1/systemone. Priced on input tokens: the state plus your questions. The short typed answers they return carry no output charge on any decision model today.

Model	Context	Input EGP / 1M
Jared Palmer: Kev 4Bjaredpalmer	8K	2.47
TypeSafe: Jev 1.13typesafe	262K	2.47
OpenAI: GPT-6 Luna DecisionsOpenAI	1M	5.89

How billing works

Text models are billed per million tokens, with input and output priced separately; image models are billed per generated image. You pay only for what a request actually uses. The rate shown is for the model's primary route; if a fallback provider serves a request, it is billed at that provider's rate, which can differ. Some models also price cached, audio, image or video tokens separately; those rates are on the model's page in the dashboard. Each model page shows its context window and modalities. For request formats and limits, see the documentation; to filter and compare models, use the model library.

How to pay from Egypt

You add credit to a prepaid EGP balance from Billing in the dashboard, and every request is paid from that balance. Checkout runs on Kashier's hosted payment page and is charged in Egyptian pounds, so you can pay with a local card or a mobile wallet. We do not store your full card details or the mobile wallet number you pay with. Refunds are covered by the refund policy.

Which cards can I pay with?: Visa, Mastercard and Meeza cards, and Apple Pay. Every payment is charged in Egyptian pounds.
Which mobile wallets can I pay with?: Vodafone Cash, Etisalat Cash, Orange Cash and WE Pay.
How fast does my credit arrive?: Credit is added to your balance once Kashier confirms the payment as successful, normally within a few seconds of checkout.