One OpenAI-compatible API for chat, image, speech and transcription.
One key, one balance.
CloudGPU's AI API is one OpenAI-compatible endpoint, https://cloudgpu.app/v1, for chat, image, speech and transcription models. Point the OpenAI SDK, Cursor or Cline at it by changing base_url, and pay per use from the same prepaid balance as GPU rentals. The table below lists the current rate for every model.
CloudGPU is in public beta. There is no SLA: no uptime guarantee and no service credits. Unused GPU minutes go back to your balance when you destroy the instance, and the balance does not expire.
Alipay, card or USDT, no Chinese phone
Alipay, Google Pay, Apple Pay, Visa or Mastercard from $5, or USDT on TRC-20 — no Chinese phone number or ID needed. The way in when your bank keeps declining the vendor's own checkout. What you pay is what lands in your balance.
One balance, many models
Claude, DeepSeek, GLM, Kimi, MiniMax and more on one prepaid balance — one bill, no juggling accounts across upstream providers.
OpenAI-compatible
Drop into the OpenAI SDK. Change base_url and key — done. Works with LangChain, AutoGen, CrewAI, OpenClaw, Cline, Continue.
Shared with GPU rental
One balance covers GPU rentals and API calls.
What every model costs
Chat, image generation, speech synthesis and transcription, all on one balance — the same balance GPU rentals draw from.
Chat & completion
per 1M tokens/v1/chat/completions| Model | Input | Output | Cached input |
|---|---|---|---|
glm-5.3-flash Zhipu GLM | $0.135 | $0.45 | $0.027 |
glm-5.3 Zhipu GLM | $1.26 | $3.96 | $0.234 |
glm-5.2 Zhipu GLM | $1.26 | $3.96 | $0.23 |
glm-5.1 Zhipu GLM | $1.26 | $3.96 | $0.24 |
kimi-k3 Moonshot Kimi | $2.8642 | $14.3212 | $0.29 |
kimi-k2.7-code Moonshot Kimi | $0.90 | $3.80 | $0.20 |
kimi-k2.6 Moonshot Kimi | $0.855 | $3.60 | $0.144 |
kimi-k2.5 Moonshot Kimi | $0.855 | $3.60 | $0.15 |
qwen3.8-max Qwen | $1.80 | $5.40 | $0.36 |
qwen3.7-max Qwen | $2.25 | $6.75 | $0.45 |
qwen3.6-plus Qwen | $0.45 | $2.70 | $0.09 |
qwen3.8-flash Qwen | $0.135 | $0.423 | $0.027 |
qwen3.5-397b Qwen | $0.54 | $3.24 | $0.30 |
qwen3.6-35b Qwen | $0.3375 | $2.025 | $0.15 |
qwen3.8-27b Qwen | $0.45 | $2.70 | $0.09 |
minimax-m3 MiniMax | $0.2814 | $1.1055 | $0.1407 |
deepseek-v4-flash DeepSeek | $0.27 | $1.08 | $0.027 |
deepseek-v4.1-flash DeepSeek | $0.27 | $1.08 | $0.009 |
deepseek-v4-pro DeepSeek | $1.3065 | $3.564 | $1.3065 |
qwen3-coder-480b Qwen | $0.45 | $1.80 | $0.15 |
gpt-oss-120b OpenAI open-weight | $0.135 | $0.54 | $0.0675 |
gpt-6-astra OpenAI GPT | $6.00 | $30.00 | $0.60 |
gpt-6-sol OpenAI GPT | $1.20 | $6.00 | $0.12 |
gpt-6.1-sol OpenAI GPT | $1.20 | $6.00 | $0.06 |
gpt-5.6-terra OpenAI GPT | $1.50 | $9.00 | $0.15 |
gpt-5.6-sol OpenAI GPT | $3.00 | $18.00 | $0.30 |
| Claude | Input | Output | Cached input | Official list price, per 1M tokens |
|---|---|---|---|---|
claude-opus-5-5 Claude | $2.00 | $10.00 | $0.10 | input $4.00 / output $20.00 / cache hits $0.20 |
claude-opus-5 Claude | $2.50 | $12.50 | $0.25 | input $5.00 / output $25.00 / cache hits $0.50 |
claude-opus-4-8 Claude | $2.50 | $12.50 | $0.25 | input $5.00 / output $25.00 / cache hits $0.50 |
claude-opus-4-7 Claude | $2.50 | $12.50 | $0.25 | input $5.00 / output $25.00 / cache hits $0.50 |
claude-opus-4-6 Claude | $2.50 | $12.50 | $0.25 | input $5.00 / output $25.00 / cache hits $0.50 |
claude-sonnet-5-5 Claude | $1.00 | $5.00 | $0.05 | input $2.00 / output $10.00 / cache hits $0.10 |
claude-sonnet-5 Claude | $1.00 | $5.00 | $0.10 | input $2.00 / output $10.00 / cache hits $0.20 |
claude-sonnet-4-6 Claude | $1.50 | $7.50 | $0.15 | input $3.00 / output $15.00 / cache hits $0.30 |
claude-fable-5-1 Claude | $5.00 | $25.00 | $0.125 | input $10.00 / output $50.00 / cache hits $0.25 |
claude-fable-5 Claude | $5.00 | $25.00 | $0.50 | input $10.00 / output $50.00 / cache hits $1.00 |
| Official list prices checked 2026-10-24 on the official Claude pricing page. They can change after that date; the official page is the source for them. CloudGPU is an independent service, not the official Claude API. | ||||
claude-opus-5-5
CloudGPU, per 1M tokens
input $2.00 / output $10.00 / cached input $0.10
Official list price, per 1M tokens
input $4.00 / output $20.00 / cache hits $0.20
claude-opus-5
CloudGPU, per 1M tokens
input $2.50 / output $12.50 / cached input $0.25
Official list price, per 1M tokens
input $5.00 / output $25.00 / cache hits $0.50
claude-opus-4-8
CloudGPU, per 1M tokens
input $2.50 / output $12.50 / cached input $0.25
Official list price, per 1M tokens
input $5.00 / output $25.00 / cache hits $0.50
claude-opus-4-7
CloudGPU, per 1M tokens
input $2.50 / output $12.50 / cached input $0.25
Official list price, per 1M tokens
input $5.00 / output $25.00 / cache hits $0.50
claude-opus-4-6
CloudGPU, per 1M tokens
input $2.50 / output $12.50 / cached input $0.25
Official list price, per 1M tokens
input $5.00 / output $25.00 / cache hits $0.50
claude-sonnet-5-5
CloudGPU, per 1M tokens
input $1.00 / output $5.00 / cached input $0.05
Official list price, per 1M tokens
input $2.00 / output $10.00 / cache hits $0.10
claude-sonnet-5
CloudGPU, per 1M tokens
input $1.00 / output $5.00 / cached input $0.10
Official list price, per 1M tokens
input $2.00 / output $10.00 / cache hits $0.20
claude-sonnet-4-6
CloudGPU, per 1M tokens
input $1.50 / output $7.50 / cached input $0.15
Official list price, per 1M tokens
input $3.00 / output $15.00 / cache hits $0.30
claude-fable-5-1
CloudGPU, per 1M tokens
input $5.00 / output $25.00 / cached input $0.125
Official list price, per 1M tokens
input $10.00 / output $50.00 / cache hits $0.25
claude-fable-5
CloudGPU, per 1M tokens
input $5.00 / output $25.00 / cached input $0.50
Official list price, per 1M tokens
input $10.00 / output $50.00 / cache hits $1.00
Official list prices checked 2026-10-24 on the official Claude pricing page. They can change after that date; the official page is the source for them. CloudGPU is an independent service, not the official Claude API.
Image generation
per image/v1/images/generations| Model | Price |
|---|---|
flux-1-schnell Black Forest Labs FLUX | $0.0015 |
flux-1-dev Black Forest Labs FLUX | $0.0125 |
flux-2-pro Black Forest Labs FLUX | $0.024 |
flux-2-dev Black Forest Labs FLUX | $0.015 |
seedream-5-flash ByteDance Seedream | $0.0175 |
Text to speech
per 1M characters/v1/audio/speech| Model | Price |
|---|---|
kokoro-tts Kokoro | $7.50 |
chatterbox-tts Chatterbox | $7.50 |
Speech to text
per second of audio/v1/audio/transcriptions| Model | Price | Per hour of audio |
|---|---|---|
whisper-v3-turbo OpenAI Whisper | $0.00005 | $0.18 |
qwen3-asr Qwen | $0.00003 | $0.108 |
whisper-v3 OpenAI Whisper | $0.00005 | $0.18 |
All prices in USD, billed from the same balance as GPU rentals.
Cached input is charged automatically when the upstream cache fires. Where it equals the input price, the upstream gives us no cache discount, so there is none to pass on.
What a month costs here
Four workload shapes we see in agent and IDE traffic, priced on deepseek-v4-flash.
| Workload | Monthly tokens | Here |
|---|---|---|
IDE code completion Cursor, Continue.dev, Tabby — inline autocomplete | 18M in + 2M out | $7.02 |
RAG production API LangChain, LlamaIndex, Haystack with 6K-token context | 26M in + 2M out | $9.18 |
Multi-agent code task AutoGen, CrewAI, OpenClaw — agents bouncing per feature | 9M in + 2M out | $4.59 |
Autonomous coding agent Cline, aider, terminal agents reading whole repos | 7M in + 1M out | $2.97 |
Computed from the table above, not typed in — if a price moves, this moves with it. Cache hits make the real bill lower.
Open-source starter · MIT
Deploy a WhatsApp Business autoresponder
Production-ready Node.js bot — Meta WhatsApp Cloud API + DeepSeek-V4 via cloudgpu. Editable faq.yml for the shop owner. One-click deploy to Railway or Fly.io.
Built for small businesses whose AI subscriptions keep getting declined at checkout — anywhere WhatsApp is the primary customer-comms channel.
5-minute quick start
curl https://cloudgpu.app/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer cgw-sk-..." \
-d '{
"model": "deepseek-v4-flash",
"messages": [{"role": "user", "content": "Hello"}]
}'smart_toyLet your AI coding assistant rent the GPU
The CloudGPU MCP server lets Cursor, Cline, opencode, Claude Code and other MCP clients quote a price, rent a GPU once you approve, run your job, download the results and shut the GPU down. CloudGPU's servers enforce the budget and time limit on every instance.
Set up the MCP server →Common questions
Why use CloudGPU instead of an account at each model vendor?
One key and one prepaid balance cover every model in the table, plus GPU rentals. You can pay by Alipay, Google Pay, Apple Pay, card or USDT.
How is my data handled?
AI API requests pass through our gateway only so they can be forwarded to the model vendor or provider serving the model you choose; we do not keep a copy of their content or of the responses. For billing we store a usage record for each request (for example time, model, token counts and amount charged); these records do not contain your prompts, the audio or images you send, or the model's responses. Prompts sent to the image-generation endpoint are also checked by the content-moderation service of our card payment processor. The vendor or provider may be located in a different country from yours, and how long it keeps request data is governed by its own terms. Full details are in our Privacy Policy.
What payment methods do you accept?
Alipay, Google Pay, Apple Pay, Visa and Mastercard in one checkout, or USDT on TRC-20, credited 1:1 in USD with no fee on top. Requests then deduct from that balance. Bank transfer is also available for business customers. Minimums for each method are on https://cloudgpu.app/payments.
What about rate limits?
100 requests per minute per API key, 10,000 requests per day per account. These are anti-abuse defaults for the public beta and can be raised on request.
Will the OpenAI SDK work?
Yes. Use the OpenAI Python or Node.js SDK and override base_url to https://cloudgpu.app/v1. The /v1/chat/completions surface is request-compatible. Claude Messages-format clients can set their base URL to https://cloudgpu.app and call /v1/messages.
Pay as you go from one prepaid balance
No credit card. The same balance covers GPU rentals.
Create your API key