# Examples of api config below

# Proprietary models examples
gpt-4-1106-preview:
    model: gpt-4-1106-preview
    endpoints: null
    api_type: openai
    parallel: 32
    max_tokens: 4096
    temperature: 0.0

gpt-4-turbo-2024-04-09:
    model: gpt-4-turbo-2024-04-09
    endpoints: null
    api_type: openai
    parallel: 32
    max_tokens: 4096
    temperature: 0.0

gpt-4.1:
    model: gpt-4.1
    endpoints: null
    api_type: openai
    parallel: 64
    max_tokens: 32000
    temperature: 0.0

gpt-4.1-mini:
    model: gpt-4.1-mini
    endpoints: null
    api_type: openai
    parallel: 64
    max_tokens: 32000

# gemini-2.5:
#     model: gemini-2.5-pro-preview-03-25
#     endpoints: null
#     api_type: gemini
#     parallel: 50

gemini-2.5:
    model: gemini-2.5-pro-exp-03-25
    endpoints: null
    api_type: vertex
    parallel: 16
    project_id: gen-lang-client-0966014990
    regions: us-central1

claude-3-7-sonnet-20250219-thinking-16k:
    model: claude-3-7-sonnet-20250219
    endpoints: null
    max_tokens: 20000
    budget_tokens: 16000
    api_type: anthropic_thinking
    parallel: 32

deepseek-r1:
    endpoints:
        - api_key: <your_api_key>
    api_type: deepseek_reasoner
    parallel: 32

claude-3-5-sonnet-20241022:
    model: claude-3-5-sonnet-20241022
    endpoints: null
    max_tokens: 20000
    api_type: anthropic
    parallel: 32

o3-mini-2025-01-31-high:
    model: o3-mini-2025-01-31
    endpoints: null
    reasoning_effort: high
    api_type: openai_thinking
    parallel: 32

gpt-4o-mini-2024-07-18:
    model: gpt-4o-mini
    endpoints: null
    api_type: openai
    parallel: 128
    max_tokens: 8196
    temperature: 0.0

# Local inference examples
qwq-32b:
    model: Qwen/QwQ-32B
    endpoints: null
    api_type: sglang
    local_engine: True
    temperature: 0.6
    end_think_token: "</think>"
    max_tokens: 32000
    
gemma-3-27b-it:
    model: google/gemma-3-27b-it
    endpoints:
        - api_base: http://0.0.0.0:<port_number>/v1
          api_key: '-'
    api_type: openai
    parallel: 128
    max_tokens: 8196
    temperature: 0.0