Supported LLMs,one interface.Pay in Korean won.
Text · Image · Audio · Video— one key.Better pricing, higher availability, domestic processing option included.Same OpenAI SDK — just change one line: base_url.
325
Models
24
Providers
99.9%
Uptime target
₩1,345
Today's rate (USD)
Get started in minutes
Just 3 steps from sign-up to your first API call.
Sign up
Sign up with email. No credit card required.
Sign up 02Top up credit
Top up credits. From 5,000 credits, with an optional monthly plan.
View pricing 03Issue an API key
Get your plm_ key and change only base_url — you're ready to call.
Quickstartfrom openai import OpenAI
client = OpenAI(
api_key="plm_xxxxxxxxxxxxxxxx",
base_url="https://apirouter.pleum.ai/v1",
)
response = client.chat.completions.create(
model="gpt-4.1",
messages=[{"role": "user", "content": "Why is the sky blue?"}],
)
print(response.choices[0].message.content)Connect your agent in one line
Claude Code, Aider, Codex — no per-tool setup, pleum launchconnect instantly.
Install
$ curl -fsSL https://router.pleum.ai/install | sh
Or run without installing: npx pleumrouter login
Requires Node.js 18 or later.
Claude Code
pleum launch claude
Aider
pleum launch aider
Codex
pleum launch codex
Goose
pleum launch goose
OpenHands
pleum launch openhands
$ npx pleumrouter login
✓ Logged in.
$ pleum launch claude
Claude Code started
Why PleumRouter
A supported model catalog with operations and billing for Korean businesses.
One API, supported models and modalities
Use one OpenAI-compatible API for supported GPT, Claude, Gemini, Grok, and domestic models. Supported image, audio, and video generation models use the same key.
Higher availability
Automatically switches to another available provider during an outage to reduce interruptions.
Pricing and performance
See the KRW cost and applied exchange rate for every response. All fees are disclosed upfront.
PII masking
When sensitive-data masking is enabled, prompts are masked before they reach providers.
Model pricing (KRW)
Today: 1 USD = ₩1,345
1050K · OpenAI
Input (credits/1M)
₩13,512
1000K · Pleum
Input (credits/1M)
₩0
1049K · Moonshot AI
Input (credits/1M)
₩4,054
131K · Upstage
Input (credits/1M)
₩203
1000K · Anthropic
Input (credits/1M)
₩13,512
262K · FriendliAI
Input (credits/1M)
₩1,621
1000K · Google
Input (credits/1M)
₩676
500K · xAI
Input (credits/1M)
₩2,702
262K · DeepInfra
Input (credits/1M)
₩115
1040K · DeepSeek
Input (credits/1M)
₩297
256K · BytePlus ModelArk
Input (credits/1M)
₩338
262K · Mistral AI
Input (credits/1M)
₩2,027
| Model | Provider | Context | Weekly usage | Input (credits/1M) | Output (credits/1M) |
|---|---|---|---|---|---|
GPT-6 Astra | OpenAI | 1050K | — | ₩13,512 | ₩67,561 |
Pleum Benchmark Smart | Pleum | 1000K | — | ₩0 | ₩0 |
Kimi K3 | Moonshot AI | 1049K | — | ₩4,054 | ₩20,268 |
Solar Pro 3 | Upstage | 131K | — | ₩203 | ₩811 |
Claude Fable 5.1 | Anthropic | 1000K | — | ₩13,512 | ₩67,561 |
EXAONE 2.0 (K-EXAONE 750B) | FriendliAI | 262K | — | ₩1,621 | ₩6,486 |
Gemini 3 Flash (Preview) | 1000K | — | ₩676 | ₩4,054 | |
Grok 4.6 | xAI | 500K | — | ₩2,702 | ₩8,107 |
NVIDIA Nemotron 3 Super 120B | DeepInfra | 262K | — | ₩115 | ₩540 |
DeepSeek V4 Flash Vision Experimental | DeepSeek | 1040K | — | ₩297 | ₩892 |
ByteDance Seed 2.0 Lite | BytePlus ModelArk | 256K | — | ₩338 | ₩2,702 |
Mistral Medium 3.5 | Mistral AI | 262K | — | ₩2,027 | ₩10,134 |
Supported providers
GPT-6 Astra
GPT-5.6 Sol
Pleum Benchmark Smart
Pleum Perfect
Kimi K3
Kimi K2.6
Kimi K3
Mistral Medium 3.5
Solar Pro 3
Solar Pro 4
Claude Fable 5.1
Claude Opus 5
Claude Sonnet 5
Claude Fable 5
EXAONE 2.0 (K-EXAONE 750B)
DeepSeek-V3.2
Gemini 3 Flash (Preview)
Gemini 3.8 Flash
Grok 4.6
Grok 4.5
DeepSeek V4 Flash Vision Experimental
DeepSeek V4 Pro
ByteDance Seed 2.0 Lite
ByteDance Seed 1.6 (250915)
Mistral Medium 3.5
Mistral Large 3
DeepSeek V4 Pro
Qwen3.8 Max
GPT-OSS 120B
GPT-OSS 20B
GPT-OSS 120B
GPT-OSS 20B
GLM-5.3 Flash
GLM-5.3
Qwen3.8 27B
GLM-5.3
Hermes 4 70B
Hermes 4 405B
MiniMax M3
MiniMax-M2.5
Kling V3
Kling V3 Turbo
LTX 2.3 Fast
LTX 2.3 Pro
ElevenLabs Multilingual v2
ElevenLabs v3
Fish Audio S2.1 Pro
Fish Audio S1
GPT-6 Astra
GPT-5.6 Sol
Pleum Benchmark Smart
Pleum Perfect
Kimi K3
Kimi K2.6
Kimi K3
Mistral Medium 3.5
Solar Pro 3
Solar Pro 4
Claude Fable 5.1
Claude Opus 5
Claude Sonnet 5
Claude Fable 5
EXAONE 2.0 (K-EXAONE 750B)
DeepSeek-V3.2
Gemini 3 Flash (Preview)
Gemini 3.8 Flash
Grok 4.6
Grok 4.5
DeepSeek V4 Flash Vision Experimental
DeepSeek V4 Pro
ByteDance Seed 2.0 Lite
ByteDance Seed 1.6 (250915)
Mistral Medium 3.5
Mistral Large 3
DeepSeek V4 Pro
Qwen3.8 Max
GPT-OSS 120B
GPT-OSS 20B
GPT-OSS 120B
GPT-OSS 20B
GLM-5.3 Flash
GLM-5.3
Qwen3.8 27B
GLM-5.3
Hermes 4 70B
Hermes 4 405B
MiniMax M3
MiniMax-M2.5
Kling V3
Kling V3 Turbo
LTX 2.3 Fast
LTX 2.3 Pro
ElevenLabs Multilingual v2
ElevenLabs v3
Fish Audio S2.1 Pro
Fish Audio S1
PleumRouter vs OpenRouter
| Feature | PleumRouter | OpenRouter |
|---|---|---|
| Payment currency | KRW ₩ | USD $ |
| Tax invoice | Coming soon | Not available |
| Payment method | KRW card payment | Foreign card only |
| Business verification | NTS business verification | Not supported |
| PII masking (PIPA) | PII masking | Not supported |
| Domestic social login | Kakao · Naver | Not supported |
| Catalog size | Focused on major models | Broader |
| Non-Korean market billing | Japan and Brazil in progress | Worldwide USD billing |
* Korea Eximbank rates are applied daily. Prices vary by model.
Frequently asked questions
Still have questions? Check the docs or reach out anytime.
Is there a Korean alternative to OpenRouter?
PleumRouter fills that role. It routes many models — OpenAI, Anthropic, Google and others — through a single API key, the same as OpenRouter. What differs is KRW billing, top-ups with Korean cards, and domestic requirements such as business verification and PII masking. It is OpenAI SDK compatible, so you only change base_url and the key.
Can I pay in Korean won?
Yes. Balances and prices are shown in credits (1 credit = KRW 1). We support Korean cards and bank transfers, so there's no USD conversion or foreign transaction fee.
Is there a markup on token prices?
No. Token prices carry no markup (0%) — you pay the provider's cost as-is. Separately, a 3% top-up fee applies when you add credits. School email (.ac.kr/.edu) accounts can use the student plan (₩19,900/month): the first 6 months are free and the fee stays 0% afterwards.
Do you issue tax invoices?
Electronic tax invoices for business accounts are in preparation. You can complete National Tax Service business verification now, and it will apply as soon as issuance begins.
How is personal data (PII) protected?
We provide sensitive-data masking to support PIPA-aligned handling. When enabled, prompts are masked before they reach providers, and you can control it per API key.
What happens if a provider goes down?
During an outage, we automatically switch to another available provider, and calls that ultimately fail are never charged.
Can I get a refund?
Unused paid credits are refundable. See the refund policy page for the full terms.
View refund policyThe answers above are current as of 2026-09-02. See pricing & fees policy · privacy policy
Running a business on PleumRouter?
Volume pricing, invoiced billing with tax invoices, partnerships — tell us how your company uses PleumRouter and our team will get back to you.
Start now
Earn credit as you go, from sign-up to first top-up · 10% bonus on your first payment · 0% token markup · Top-up fee 3% · Student plan: 6 months free
Start for free