Use 319 AI models with one API key
Native OpenAI, Anthropic and Gemini APIs. Point Claude Code or Codex at a new base URL and go. With TEE proof on, you can check which provider handled your requests.
Newly added
The latest models to go live.
- -90%
- Input
- from $0.80list $8.00
- Output
- from $4.00list $40.00
- Context
- 272K
- Max output
- 128K
- Capability
- —
AcceptsTextImageGPT-6.2 Astra for Codex (Responses API).
- -10%
- Input
- from $7.20list $8.00
- Output
- from $36.00list $40.00
- Context
- 1.1M
- Max output
- 128K
- Capability
- —
AcceptsTextImagePDFOpenAI's most capable model for the hardest projects, from complex coding and computer use to finished documents, slides, and spreadsheets.
- -90%
- Input
- $0.80list $8.00
- Output
- $4.00list $40.00
- Context
- 200K
- Max output
- 64K
- Capability
- —
AcceptsTextImagePDFClaude Fable 5.2 for Claude Code, with 200K context.
- Input
- $0.33
- Output
- $1.10
- Context
- 1M
- Max output
- 1M
- Capability
- —
AcceptsTextImageThinking Machines Lab's open-weight 594B multimodal MoE with a 1M-token context and adjustable reasoning effort.
Strongest from each lab
Each lab’s highest-scoring model on Epoch AI’s ECI, strongest first.
Claude Fable 5.2
#1Ranked 1 by capability- Input
- from $8.80
- Output
- from $44.00
- Context
- 1M
- Max output
- 128K
- Capability
- 173
AcceptsTextImagePDFAnthropic's most capable widely released model for demanding reasoning and long-horizon agentic work: agent sessions that run for hours, multistep deep research, and analysis carried through to a finished document.
Gemini 4 Carbon
#2Ranked 2 by capability-10%- Input
- from $1.80list $2.00
- Output
- from $9.00list $10.00
- Context
- 1M
- Max output
- 66K
- Capability
- 170
AcceptsTextImageAudioVideoPDFGoogle's frontier Gemini 4 model for the hardest reasoning, coding, and security work.
GPT-6 Astra
#3Ranked 3 by capability-10%- Input
- from $9.00list $10.00
- Output
- from $45.00list $50.00
- Context
- 1.1M
- Max output
- 128K
- Capability
- 169
AcceptsTextImagePDFOpenAI's most capable model, built for the hardest end-to-end work: complex reasoning, coding, computer use, research, and document creation.
Kimi K3.1 Pro
#15Ranked 15 by capability- Input
- $2.0878
- Output
- $10.9208
- Context
- 1M
- Max output
- 1M
- Capability
- 161
AcceptsTextImageVideoMoonshot AI's flagship 3T-class open-weight model with 1M context, strong coding, 3D creation, and financial research.
Cheapest LLM APIs
Paid models by price per 1M tokens, lowest first.
| Model | Input | Output | Context | Capability |
|---|---|---|---|---|
Claude Haiku 5.5 (Reseller)-90%claudecode | $0.01 | $0.05 | 200K | — |
GPT-6 Luna (Reseller)-90%codex | $0.01 | $0.05 | 272K | — |
Codex Auto-review (Reseller)-90%codex | $0.01 | $0.05 | 272K | — |
GPT-5.6 Luna (Reseller)-90%codex | $0.02 | $0.12 | 1.1M | — |
GLM-4.6V-FlashXZ.ai | $0.024092 | $0.2409 | 128K | — |
Longest context window models
Largest context window first.
| Model | Context | Input / Output | Capability |
|---|---|---|---|
| 2M | $1.375 / $2.75 | — | |
| 2M | $1.375 / $2.75 | — | |
| 2M | $1.375 / $2.75 | — | |
GPT-6.2 Astra-10%OpenAI | 1.1M | $7.20 / $36.00 | — |
GPT-6.2 Sol-10%OpenAI | 1.1M | $1.80 / $9.00 | — |
Cheaper than official prices
Why OhMyGPT
- TEE proof of which provider answered
- With TEE proof on for a key, responses carry a signed receipt you can verify, recording the provider host (such as api.anthropic.com) and what it returned. Learn more
- Automatic failover
- Each model has several providers. If one errors or is rate limited, the request moves to the next before any output starts. Provider uptime is public. Learn more
- Billed on actual usage
- You pay for the usage the provider reports, with cache priced separately. Failed requests are free, and every request has an itemized receipt. Learn more
- Native APIs, no code changes
- Native OpenAI Chat Completions and Responses, Anthropic Messages and Gemini APIs. Claude Code and Codex connect by changing the base URL. Learn more
- Your providers, your order
- Pick which providers to use per request, set their order, or exclude one. If you don't choose, OhMyGPT picks for you and fails over to the next. Learn more
- Payment and invoices
- Pay by card, Alipay and more. Invoices and receipts are available.
Latest announcements
Quick start
curl https://api.ohmygpt.com/v1/chat/completions \
-H "Authorization: Bearer sk-…" \
-H "Content-Type: application/json" \
-d '{"model": "openai/gpt-oss-120b", "messages": [{"role": "user", "content": "Hello"}]}'