Live on staging

One endpoint.
Every model.

Access Claude, GPT, Gemini, and DeepSeek through a single API. Supports both OpenAI and Anthropic request formats.

20
Models
4
Providers
2
API Formats
curl # One URL. Any model. That's it.

curl https://api.anannas.ai/v1/chat/completions \
  -H "Authorization: Bearer an-sk-..." \
  -d '{
    "model": "claude-sonnet-4-20250514",
    "messages": [{"role": "user", "content": "Hello"}]
  }'
✓ OpenAI format ✓ Anthropic format ✓ Any HTTP client
⚡

One endpoint

One base URL for all providers. Switch models by changing a string — no config, no SDK changes, no rewrites.

🔀

Both standards

Send requests in OpenAI or Anthropic format. Use whichever SDK you already have, or just curl it.

📡

Streaming

Full SSE streaming from every provider. All normalized to a consistent chunk format.

🔧

Tool calling

Send tools once in OpenAI format. Anannas translates to Anthropic, Google, and DeepSeek schemas automatically.

🧠

Thinking & reasoning

Extended thinking, reasoning_effort, budget_tokens — works across providers with automatic translation.

🔑

Bring your own keys

Use our managed keys or plug in your own provider API keys. BYOK runs at zero markup.

Models

20 models across 4 providers. All accessible through the same endpoint, same format.

Anthropic6 models
claude-opus-4-6-20260205
1M ctx$5 / $25
claude-sonnet-4-6-20260217
1M ctx$3 / $15
claude-opus-4-20250514
200K ctx$15 / $75
claude-sonnet-4-20250514
200K ctx$3 / $15
claude-3-5-sonnet-20241022
200K ctx$3 / $15
claude-3-5-haiku-20241022
200K ctx$0.80 / $4
OpenAI9 models
gpt-4.1
1M ctx$2 / $8
gpt-4.1-mini
1M ctx$0.40 / $1.60
gpt-4.1-nano
1M ctx$0.10 / $0.40
gpt-4o
128K ctx$2.50 / $10
gpt-4o-mini
128K ctx$0.15 / $0.60
o3
200K ctx$2 / $8
o4-mini
200K ctx$1.10 / $4.40
gpt-5
1M ctx$2 / $8
gpt-5-mini
1M ctx$0.80 / $3.20
Google3 models
gemini-2.5-pro
1M ctx$1.25 / $10
gemini-2.5-flash
1M ctx$0.15 / $0.60
gemini-2.5-flash-lite
1M ctx$0.075 / $0.30
DeepSeek2 models
deepseek-chat
128K ctx$0.27 / $1.10
deepseek-reasoner
128K ctx$0.55 / $2.19

Prices shown as input / output per 1M tokens. Anannas adds a 5% markup on managed keys.

API

Two formats, same models. Use whichever you already know.

EndpointFormatDescription
POST /v1/chat/completions OpenAI Chat completions — streaming & non-streaming
POST /v1/messages Anthropic Messages — same models, Anthropic request format
GET /v1/models List all available models
GET /v1/models/:id Get model details and pricing

Pricing

5% markup on provider costs

No per-request fees. No monthly minimum. No seat charges. You pay what you use.

0%

BYOK — bring your own keys