One key for every model you run.

Point Claude Code, OpenCode, or any OpenAI SDK at Inferio. 23 live models. $0.01 a call.

Five labs. Pick by job, not by brochure.

xAI, Zhipu, DeepSeek, OpenAI, iFlytek. Cycles the jobs. Same $0.01 call.

xAI · Zhipu · DeepSeek

Coding

Claude Code, OpenCode, Codex. Point the client. Keep the editor.

grok-build-0.1deepseek-v4-flashglm-5.3-flashgrok-4.6grok-composer-2.5-fast

Same client. This base URL.

OpenAI SDKs take /v1. Claude Code takes the host, no suffix.

themapi.openai.com
export OPENAI_BASE_URL=https://api.openai.com/v1
export OPENAI_API_KEY=sk-...
usapi.inferio.net
export OPENAI_BASE_URL=https://api.inferio.net/v1
export OPENAI_API_KEY=YOUR_API_KEY
same curlmodel grok-4.6
curl -X POST $OPENAI_BASE_URL/chat/completions \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"grok-4.6","messages":[{"role":"user","content":"Hello!"}],"stream":true}'
~/.claude/settings.jsonno /v1
{
  "env": {
    "ANTHROPIC_BASE_URL": "https://api.inferio.net",
    "ANTHROPIC_API_KEY": "YOUR_API_KEY",
    "MAX_THINKING_TOKENS": "0",
    "CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS": "1",
    "CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC": "1"
  }
}
~/.codex/config.tomlwire_api = responses
# ~/.codex/config.toml
model = "grok-4.6"
model_provider = "inferio"

[model_providers.inferio]
name = "Inferio"
base_url = "https://api.inferio.net/v1"
wire_api = "responses"
env_key = "OPENAI_API_KEY"
librechat.yamlmodels.fetch
endpoints:
  custom:
    - name: 'Inferio'
      apiKey: '${OPENAI_API_KEY}'
      baseURL: 'https://api.inferio.net/v1'
      models:
        fetch: true
      titleConvo: true
      modelDisplayLabel: 'Inferio'
All 29 setup guides

After the key: spend, requests, keys.

Spend, requests, keys. The account after you mint one.

Dashboard

Good afternoon, Exits

Monthly spend

$12.47

8.2% vs prior window

Requests

1,247 calls

Tokens

4.2M logged

PerformanceRPM 2.14status healthy
  • grok-4.641%
  • glm-5.3-flash28%
  • grok-chat-fast18%
  • deepseek-v4-flash13%
Recommended
  • Credit is healthy. Keys on /token. Logs on /logs.
  • claude-code · opencode · sillytavern · ci-runner
Open dashboard

every live id

$0.01

A call. Not a million tokens. grok-4.6, glm-5.3-flash, grok-imagine-image-edit, grok-stt: same rate.

$5

500

calls

$20

2,000

calls

$100

10,000

calls

$20

2,000 calls

Pricing page

Keep the client. Swap the key.

29 setup guides. OpenAI shape on /v1. Claude Code takes the host, no suffix. Catalog is live. Support is [email protected].

LibreChat Integration GuideOpen WebUI Integration GuideLobeChat Integration GuideAnythingLLM Integration GuideCherry Studio Integration GuideTypingMind Integration GuideBoltAI Integration GuidePage Assist Integration GuideChatbox Integration GuideBig-AGI Integration GuideSillyTavern Integration GuideJanitor.AI Integration GuideRisuAI Integration GuideChub / Venus Integration GuideNevika Integration GuideOpenCode Setup GuideKilo Code Setup GuideZed IDE Setup GuideCline Setup GuideRoo Code Setup GuideContinue.dev Setup GuideAider Setup GuideCC Switch Integration GuideClaude Code IntegrationOpenAI Codex CLI Setup GuideGemini CLI IntegrationOpenClaw IntegrationHermes Agent IntegrationMCP ClientsLibreChat Integration GuideOpen WebUI Integration GuideLobeChat Integration GuideAnythingLLM Integration GuideCherry Studio Integration GuideTypingMind Integration GuideBoltAI Integration GuidePage Assist Integration GuideChatbox Integration GuideBig-AGI Integration GuideSillyTavern Integration GuideJanitor.AI Integration GuideRisuAI Integration GuideChub / Venus Integration GuideNevika Integration GuideOpenCode Setup GuideKilo Code Setup GuideZed IDE Setup GuideCline Setup GuideRoo Code Setup GuideContinue.dev Setup GuideAider Setup GuideCC Switch Integration GuideClaude Code IntegrationOpenAI Codex CLI Setup GuideGemini CLI IntegrationOpenClaw IntegrationHermes Agent IntegrationMCP Clients

Put the key in the client. Keep the rest of your stack.

23 ids, $0.01 a call. Support is [email protected].