Your coding agents,
without the meter.
Claude Code, Codex, Gemini CLI and more on 27 frontier models. Unlimited tokens for one flat monthly price — so your agents finish the job instead of stopping at a budget.
3 days on Solo · card required · cancel before day 4 and you pay nothing
- Tokens this session
- 0
- Per-token price*
- $0.00
- Billed on Piramyd
- $0.00
*Illustrative session, at $3 per 1M tokens. On Piramyd it's covered by your flat plan.
Set up in one command for
Claude Code
Codex CLI
Gemini CLI
OpenCode
Kimi Code
Qwen CLI
OpenClaw
Products and teams shipping on Piramyd.
Their agents run on Piramyd every day, on one flat plan.
An assistant on your computer and phone, for one monthly price.
ÓsmioA general-purpose AI agent platform: chat, image, video and audio, web research and a code sandbox.
Helps businesses in Angola get found online.
SAP consulting, implementation and training for companies in Angola.
CharapterUnlimited tokens for roleplay on one flat subscription, for SillyTavern, JanitorAI or your own app.
Building on Piramyd?Get your product featured here.Tokens our users didn’t have to count.
Every token developers’ agents have used through Piramyd, updated as it happens. None of it billed per token.
- Input
- —
- Output
- —
Read from the public API every 30 seconds; between reads it moves at the rate last measured.
Most-used models
Models in today’s catalog, by tokens served
A meter makes you stop
your agent halfway.
Same agent, same task, two ways to pay. When every token is billed, the rational move is to cut the run short. When it isn't, the rational move is to finish.
Billed per token
List price, ~$3 per 1M tokens
This task
$0.00
0.0M tokens
Piramyd
Flat monthly plan, unlimited tokens
Extra for this task
$0.00
0.0M tokens
- Trim the context it neededGive it the whole repo
- Stop the run at 80%Let it finish and verify
- One agent at a timeReview, docs and tests in parallel
- Skip the big refactorShip the refactor
Illustrative: a 22.4M-token refactor at $3 per 1M tokens with a $40 budget alert.
From install to unlimited in about a minute.
Keep the agents you already use. Piramyd replaces the bill, not your workflow.
- 01
Start your plan
Subscribe in the dashboard and copy your API key. Solo starts with a 3-day trial.
- 02
Run npx piramyd
Paste the key, pick your agents. Each gets its own Piramyd profile; your usual setup, CLAUDE.md, hooks and memory stay as they are.
- 03
Let them work
Long refactors, parallel agents, research that reads everything. No token counter, no overage, one monthly price.
Configured for you
Anything else that speaks the OpenAI or Anthropic API works too: set the base URL to https://api.piramyd.cloud/v1.
- Claude Code
claude-piramyd - Codex CLI
codex-piramyd - Gemini CLI
- OpenCode
- Kimi Code
- Qwen CLI
- OpenClaw
27 frontier models. One API key.
Switch models per task without switching tools or bills. One API key speaks both the OpenAI and Anthropic formats.
Claude Opus 4.6
Anthropic
ScaleOpus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks.
- Reasoning
- Image input
- File input
- Tool calling
- Context
- 1M
- Output
- 128K
- Input
- Text, Images, Files
Claude Sonnet 4.6
Anthropic
Growth+Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work.
- Reasoning
- Image input
- File input
- Tool calling
- Context
- 1M
- Output
- 128K
- Input
- Text, Images, Files
GPT-6 Astra
OpenAI
ScaleGPT-6 Astra is OpenAI's flagship model for demanding end-to-end work.
- Reasoning
- Image input
- File input
- Tool calling
- Context
- 1.1M
- Output
- 128K
- Input
- Files, Images, Text
Gemini 3.1 Pro
Google
ScaleGemini 3.1 Pro is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows.
- Reasoning
- Image input
- File input
- Tool calling
- Context
- 1M
- Output
- 66K
- Input
- Audio, Files, Images, Text, Video
DeepSeek V4 Pro 0813
DeepSeek
Growth+DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. This is the GA release of DeepSeek V4 Pro.
- Reasoning
- Tool calling
- JSON output
- Context
- 1M
- Output
- 393K
- Input
- Text
GLM 5.2
Z.ai
Growth+GLM 5.2 is a large-scale reasoning model from Z.ai.
- Reasoning
- Tool calling
- JSON output
- Context
- 1M
- Output
- 944K
- Input
- Text
More than chat. One API key.
The same key and base URL also give you embeddings, speech and the web.
Embeddings
Turn text into vectors for search and RAG. Multilingual models included.
POST /v1/embeddingsSpeech-to-text
Whisper transcription for files up to 25 MB, with JSON, text, SRT or VTT output.
POST /v1/audio/transcriptions- Coming soon
Text-to-speech
Natural voices from text. Models are being added.
POST /v1/audio/speech Web search
Multi-engine search results your agents can cite.
POST /v1/searchScrape & crawl
Clean content from one page or a whole site, ready for an LLM.
POST /v1/scrape · /v1/crawl
One flat price. Unlimited tokens.
Pick a plan by the models and throughput your agents need. Every plan includes unlimited tokens and no per-token charges.
Solo
One developer and their agents.
$30/month
- Unlimited tokens, no overage
- 11 models (Solo)
- 10 requests per minute
- About 2 agents working at once
- Community support
Card required · cancel before day 4
Growth
Most popularHeavy daily use, or a small team.
$50/month
- Unlimited tokens, no overage
- 20 models (Solo + Growth)
- 25 requests per minute
- About 5 agents working at once
- Priority support
Cancel anytime
Scale
Teams running agents in parallel all day.
$90/month
- Unlimited tokens, no overage
- All 27 models (Solo + Growth + Scale)
- 50 requests per minute
- About 10 agents working at once
- Dedicated support
Cancel anytime
3-day trial on Solo · Card required · Cancel before day 4 and pay nothing
What would this cost per token?
An estimate with your numbers. Agents resend the conversation on every step, so a working hour adds up fast. Defaults are on the low side; change any of them.
88M tokens a month
- Billed per token
- $264
- On Piramyd Solo
- $30
About $234 a month less, and no reason to stop an agent early.
Compare plans
| Feature | Solo | Growth | Scale |
|---|---|---|---|
| Monthly price | $30 | $50 | $90 |
| Tokens | Unlimited | Unlimited | Unlimited |
| Per-token charges | None | None | None |
| Models | 11 (Solo) | 20 (Solo + Growth) | All 27 (Solo + Growth + Scale) |
| Requests per minute | 10 | 25 | 50 |
| Agents working at once* | ≈ 2 | ≈ 5 | ≈ 10 |
| Extra throughput | +5 RPM for $15, one-time | +5 RPM for $15, one-time | +5 RPM for $15, one-time |
| Trial | 3 days | — | — |
| Support | Community | Priority | Dedicated |
* A coding agent in an active loop sends about 5 requests a minute. Short bursts over your limit wait in line for a few seconds instead of failing.
Questions
Is it really unlimited?
Yes. There is no token quota, no monthly cap and no overage bill. The only limit is how many requests you send per minute, which is what each plan sets.
What happens when I hit my requests-per-minute limit?
Short bursts wait in line for a few seconds and then go through, so your agent doesn't see an error. If you keep sending above your limit, requests get a “retry in 5 seconds” answer, which every major agent handles on its own. You can add 5 RPM for a one-time $15 from the dashboard at any time; it stays on your account.
How does the 3-day trial work?
Start Solo with a card. You get the full plan for 3 days and aren't charged until day 4. Cancel from the dashboard before then and you pay nothing.
Which models do I get?
Each plan unlocks its tier of models: Solo includes fast everyday coding models, Growth adds deeper reasoning and frontier models, and Scale unlocks the entire catalog including the largest flagship models.
Will my agent's setup change?
No. npx piramyd gives each agent its own Piramyd profile and leaves your default configuration, instructions, hooks and memory untouched.
How is my data handled?
We don't store your prompts or outputs. Requests are processed securely and encrypted in transit. Your data is yours.
Can I cancel anytime?
Yes, from the dashboard. There is no contract and no cancellation fee.
Stop rationing your agents.
Let them finish the job.
The agents you already use, every frontier model, and a bill that doesn’t move when they work harder.
3-day trial on Solo · Card required · Cancel before day 4 and pay nothing
For AI agents
Give this link to any agent: it describes the whole API.