OpenRouter

💰 Plan & Pricing

ℹ️

Free + Paid Tiers
33 free models (suffix :free) with 3-5 req/min rate limits. 323+ paid models with per-token pricing. Best free model aggregator.

🔑 API Key

sk-or-...2de2

🌐 Endpoint

https://openrouter.ai/api/v1/chat/completions

📦 Models — 33 Free + 356 Paid

✅ Free Models (33 :free tier)

Model	Category	Notes
anthropic/claude-3.5-sonnet:free	Chat	✅ Free Claude 3.5 Sonnet
google/gemini-2.5-pro:free	Chat/Reasoning	✅ Free Gemini Pro
google/gemini-2.5-flash:free	Chat	✅ Free Gemini Flash
meta-llama/llama-3.1-8b:free	Chat	✅ Free Llama 3.1 8B
meta-llama/llama-3.1-70b:free	Chat	✅ Free Llama 3.1 70B
deepseek/deepseek-r1:free	Reasoning	✅ Free DeepSeek R1
deepseek/deepseek-v3:free	Chat	✅ Free DeepSeek V3
qwen/qwen3-32b:free	Chat	✅ Free Qwen3 32B
mistralai/mistral-7b:free	Chat	✅ Free Mistral 7B
microsoft/phi-3-mini:free	Chat	✅ Free Phi-3 Mini

+ 23 more free models — rate limited to 3-5 req/min

💰 Key Paid Models

Model	Speed	Category	Notes
anthropic/claude-opus-4	⚡ 30 tok/s	Reasoning	Claude Opus 4
anthropic/claude-sonnet-4	⚡ 32 tok/s	Chat	Claude Sonnet 4
openai/gpt-4.1	⚡ 45 tok/s	Chat	GPT-4.1
google/gemini-2.5-pro	⚡ 67 tok/s	Chat/Reasoning	🥇 Best Gemini Pro
google/gemini-2.5-flash	⚡ 74 tok/s	Chat	Good Gemini Flash
openai/o1-preview	⚡ 20 tok/s	Reasoning	OpenAI o1
deepseek/deepseek-v4	⚡ 35 tok/s	Chat	DeepSeek V4
meta-llama/llama-4-maverick	⚡ 40 tok/s	Chat	Llama 4 Maverick
qwen/qwen3-235b	⚡ 38 tok/s	Chat/Coding	Qwen3 235B

+ 347 more paid models — largest model catalog available

💻 cURL Example

curl -X POST https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer sk-or-...2de2" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemini-2.5-pro",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

🐍 Python Example

from openai import OpenAI

client = OpenAI(
    api_key="sk-or-...2de2",
    base_url="https://openrouter.ai/api/v1"
)

response = client.chat.completions.create(
    model="google/gemini-2.5-pro",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)

⚠️ Pitfalls & Notes

⚠️

Free Models Rate Limited — Free models (suffix :free) are rate limited to 3-5 requests per minute. Returns 429 when exceeded.

💡

Best Gemini Provider — OpenRouter is the best provider for Gemini 2.5 Pro at 67 tok/s and Gemini 2.5 Flash at 74 tok/s.

💡

Largest Catalog — 356 models from every major provider, including many experimental and niche models not available elsewhere.

🏷️ Categories

Chat Vision Coding Reasoning TTS Video Audio

🔀 OpenRouter