Skip to content
OhMyGPT

Search

Search pages, models and actions, or paste a request ID…

One key for 301 models

56 of them are priced below official rates. Natively compatible with OpenAI, Anthropic, and Gemini—just swap the base URL in your existing code, Claude Code, or Codex.

Why OhMyGPT

Serving trillions of tokens reliably since 2023.

  • TEE-signed receipts

    Get the exact provider you asked for. Signed responses let you verify authenticity and eliminate trust concerns—reliable enough for academic research and production workloads.

    How TEE proof works
  • Cache cost optimization

    Stop paying full price for repeated context. We continuously monitor cache hit rates and optimize routing to keep long conversations and agentic loops cost-effective.

    Learn about prompt caching
  • Native APIs

    No wrappers, no schema translation. Full native support for OpenAI, Anthropic, and Gemini APIs—just update your base URL and run.

    API reference
  • Coding agents

    Connect and start shipping in one command. Use our open-source CLI omgvibe with Claude Code and Codex without touching your existing local config.

    Set up coding agents
  • Automatic failover

    If one provider goes down or hits a rate limit, another takes over instantly. Multi-provider redundancy fails over seamlessly before the first token so your app stays online.

    Reliability
  • Provider choice

    Pin preferred providers, set fallback order, or block specific ones per request—or let our router automatically pick the best route for you.

    Provider routing
  • Payments and invoices

    Billed in USD via Stripe—pay by card, Apple Pay, or Google Pay, or use Stripe Invoices for business accounts. Download invoices and receipts anytime for easy expense reporting.

  • Enterprise plans

    Custom plans for high-volume workloads. Get tailored routing policies, dedicated high-concurrency quotas, contract pricing, and unified management for your own endpoints and cloud accounts.

    Contact sales
Fastest-growing models by user adoption this week.

Newly added

Newly added models ready for you to test.

Strongest from each lab

Each lab’s top model on Epoch AI’s ECI.
  • Claude Fable 5.2Capability173
    Inputfrom $8.80Outputfrom $44.00Context1M
  • GPT-6.2 AstraCapability173
    Inputfrom $7.20list $8.00Outputfrom $36.00list $40.0010% offContext1.1M
  • Gemini 4 CarbonCapability170
    Inputfrom $1.80list $2.00Outputfrom $9.00list $10.0010% offContext1M
  • Kimi K3.1 ProCapability161
    Input$2.0878Output$10.9208Context1M
  • DeepSeek V4.1 Flash 1017Capability159
    Inputfrom $0.0657list $0.146Outputfrom $0.2628list $0.58455% offContext1M

Cheaper than official prices

56 models are priced below official rates—here are this week's most-used picks. Shown prices are the lowest per-million-token rates across providers (strikethrough shows official pricing). Discounts use a realistic 7:2:1 (cache read : input : output) blended ratio, never cherry-picked metrics.

All models

Latest announcements

All announcements