Skip to main content
Access cutting-edge AI models from leading providers through a single, unified API. All models are accessible using their model ID in your API calls.

​
Live Models Data

The catalog below is fetched live from https://api.passy.ai/v1/models and refreshes every 30 seconds.

​
Model Selection Guide

​
By Use Case

Fast Responses

gpt-4o-mini, gemini-1.5-flash, passy/mistral-nemo, passy/llama-3.1-8b-instruct

Complex Reasoning

gpt-4o, claude-3-opus-20240229, gemini-1.5-pro, passy/deepseek-v3

Cost-Effective

gpt-4o-mini, passy/mistral-nemo, passy/llama-3.1-8b-instruct

Large Context

gemini-1.5-pro (1M), gemini-1.5-flash (1M), claude-3-opus-20240229 (200K)

Vision Tasks

gpt-4o, gpt-4o-mini, gemini-1.5-pro, gemini-1.5-flash

Code Generation

gpt-4o, claude-3-sonnet-20240229, passy/deepseek-v3

​
By Budget

Economy

Models: gpt-4o-mini, passy/mistral-nemo, passy/llama-3.1-8b-instructUse Cases: Prototyping, bulk generation, simple assistantsCost: Lowest prompt and completion pricing in the current catalog

Standard

Models: gpt-4o, passy/llama-3.1-70b-instruct, gemini-1.5-flashUse Cases: Production apps, copilots, customer supportCost: Balanced performance and pricing

Premium

Models: claude-3-opus-20240229, passy/deepseek-v3, gemini-1.5-proUse Cases: Advanced reasoning, research, long-form generationCost: Higher tier for deeper reasoning and larger contexts

Enterprise

Models: claude-3-opus-20240229, gemini-1.5-pro, gpt-4oUse Cases: Critical workflows, premium support paths, broad modality coverageCost: Premium pricing for maximum reliability and quality

​
Using Models in Code

Always use the model ID when making API calls: 
from openai import OpenAI

client = OpenAI(
    base_url="https://api.passy.ai/v1",
    api_key="your-passy-api-key"
)

for model_id in [
    "gpt-4o-mini",
    "claude-3-sonnet-20240229",
    "passy/deepseek-v3"
]:
    response = client.chat.completions.create(
        model=model_id,
        messages=[{"role": "user", "content": "Hello!"}]
    )

    print(model_id, response.choices[0].message.content)

​
Automatic Fallback

Configure automatic fallback using model IDs: 
response = client.chat.completions.create(
    model="gpt-4o",
    messages=messages,
    fallback_models=[
        "claude-3-sonnet-20240229",
        "passy/deepseek-v3",
        "gemini-1.5-flash"
    ],
    fallback_on_rate_limit=True,
    fallback_on_error=True
)

​
Reference Pricing

Sample prices below are based on the model data returned by https://api.passy.ai/v1/models:
ModelFamilyContextPrompt $/MCompletion $/M
gpt-4o-miniOpenAI128K$0.15$0.60
gpt-4oOpenAI128K$2.50$10.00
claude-3-sonnet-20240229Anthropic200K$3.00$15.00
gemini-1.5-flashGoogle1M$0.08$0.30
passy/deepseek-v3DeepSeek64K$1.00$2.00
passy/mistral-nemoMistral128K$0.10$0.10
Important: Model IDs are case-sensitive. Always use the exact model ID as returned by the live models endpoint.