Same models, same API, 20% below list price.
No subscription. No fee on credits. Change the base URL and the key, and keep your code.
- models routed
- 26
- named hosts: Anthropic, DeepInfra, Novita, OpenAI
- 4
- below list price
- 20%
- fee on credits
- 0%
What you get
Same API
Chat completions, messages, responses and models work as they do on OpenRouter. Change the base URL and the key.
https://api.plaingateway.com/Move from OpenRouterHost pinned
Each model runs on one named host. We switch hosts only when that host returns an error.
X-PG-HostHow routing worksCost on every response
Each call returns its cost in a header and in the usage block. No second lookup.
X-PG-Cost-USD, usage.costResponse headersPrivacy controls
We keep no prompts or outputs. Ask for zero retention or no data collection, and the gateway refuses a host outside that policy.
provider.zdr, provider.data_collectionData policiesMost used models
Price per 1M tokens, read 2026-10-15 00:10 UTC. OpenRouter is list price plus its 5.5% fee on credits.
- DeepSeek V4.1 Flash deepseek/deepseek-v4.1-flashours $0.12 / $0.48, OpenRouter $0.158 / $0.633
- Step 5 Preview stepfun/step-5-previewours $0.80 / $2.16, OpenRouter $1.05 / $2.85
- GLM 5.3 Flash z-ai/glm-5.3-flashours $0.12 / $0.40, OpenRouter $0.158 / $0.527
- MiMo-V2.6-Flash xiaomi/mimo-v2.6-flashours $0.112 / $0.224, OpenRouter $0.148 / $0.295
- Hy4 preview tencent/hy4-previewours $0.667 / $2.00, OpenRouter $0.88 / $2.64
| Model | plaingateway in / out | OpenRouter in / out | List in / out | vs OpenRouter |
|---|---|---|---|---|
| DeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash | $0.12 / $0.48 | $0.158 / $0.633 | $0.15 / $0.60 | 24% less |
| Step 5 Previewstepfun/step-5-preview | $0.80 / $2.16 | $1.05 / $2.85 | $1.00 / $2.70 | 24% less |
| GLM 5.3 Flashz-ai/glm-5.3-flash | $0.12 / $0.40 | $0.158 / $0.527 | $0.15 / $0.50 | 24% less |
| MiMo-V2.6-Flashxiaomi/mimo-v2.6-flash | $0.112 / $0.224 | $0.148 / $0.295 | $0.14 / $0.28 | 24% less |
| Hy4 previewtencent/hy4-preview | $0.667 / $2.00 | $0.88 / $2.64 | $0.834 / $2.50 | 24% less |
| GPT-6 Lunaopenai/gpt-6-luna | $0.08 / $0.40 | $0.105 / $0.527 | $0.10 / $0.50 | 24% less |
| Claude Opus 5.5anthropic/claude-opus-5.5 | $3.20 / $16.00 | $4.22 / $21.10 | $4.00 / $20.00 | 24% less |
| DeepSeek V4 Flash 0731deepseek/deepseek-v4-flash-0731 | $0.12 / $0.48 | $0.158 / $0.633 | $0.15 / $0.60 | 24% less |
| GLM 5.3z-ai/glm-5.3 | $1.12 / $3.52 | $1.48 / $4.64 | $1.40 / $4.40 | 24% less |
| DeepSeek V4 Flash 0423deepseek/deepseek-v4-flash | $0.12 / $0.48 | $0.158 / $0.633 | $0.15 / $0.60 | 24% less |
Works with
- AiderPair programming in the terminal.
- Anthropic SDKPython and TypeScript, Anthropic style.
- Claude CodeAnthropic's coding agent in the terminal.
- ClineCoding agent inside VS Code.
- Codex CLIOpenAI's coding agent in the terminal.
- ContinueCode assistant for VS Code and JetBrains.
- CursorCode editor with a built-in assistant.
- Kilo CodeCoding agent inside VS Code.
- LangChainFramework for chains and agents.
- LibreChatSelf-hosted chat interface.
- n8nWorkflow automation.
- Open WebUISelf-hosted chat interface.
- OpenAI SDKPython and TypeScript, OpenAI style.
- OpenCodeOpen source coding agent in the terminal.
- SillyTavernChat front end for role play.
- Vercel AI SDKTypeScript toolkit for AI apps.
For agents
Paste this into an agent's context. It holds everything the first call needs.
Base URL, OpenAI style: https://api.plaingateway.com/api/v1 Base URL, Anthropic style: https://api.plaingateway.com/api API key: sk-pg-v1-... (header: Authorization: Bearer <key>) Model: deepseek/deepseek-v4.1-flash Model list: GET https://api.plaingateway.com/api/v1/models
Scroll the block sideways to read each full line.
Refer and earn 2% of every top-up, for one year after their first top-up
Every account gets a referral code at sign-up. When someone you refer adds money, 2% lands in your balance as credit, for one year after their first top-up.