Skip to main content
Glama

MCP Sage

by jalehman
models.yaml2.36 kB
# Model Configuration for MCP Sage # This file defines all available models and their configurations # Model Registry - Define all available models here models: gpt5: name: "gpt-5-2025-08-07" type: "openai" tokenLimit: 400000 costPerInputToken: 0.00000125 # $1.25 per 1M tokens costPerOutputToken: 0.00001 # $10.00 per 1M tokens gpt41: name: "gpt-4.1-2025-04-14" type: "openai" tokenLimit: 1047576 # ~1M tokens costPerInputToken: 0.000002 # $2.00 per 1M tokens costPerOutputToken: 0.000008 # $8.00 per 1M tokens gemini25pro: name: "gemini-2.5-pro" type: "gemini" tokenLimit: 1000000 costPerInputToken: 0.000002 # $2.00 per 1M tokens (approximation) costPerOutputToken: 0.000012 # $12.00 per 1M tokens (approximation) opus41: name: "claude-opus-4-1-20250805" type: "anthropic" tokenLimit: 200000 costPerInputToken: 0.000015 # $15.00 per 1M tokens costPerOutputToken: 0.000075 # $75.00 per 1M tokens # Tool-specific configurations tools: opinion: # Models that can participate in debates debateParticipants: - gpt5 - gemini25pro # Model to use as judge in debates judgeModel: opus41 # Priority order for non-debate mode (subject to token constraints) # First model that fits token limit will be used preferredModels: - gpt5 # First choice for ≤400K tokens - gemini25pro # Second choice for ≤1M tokens - gpt41 # Fallback for ≤1M tokens review: # Models that can participate in debates debateParticipants: - gpt5 - gemini25pro # Model to use as judge in debates judgeModel: opus41 # Priority order for non-debate mode (subject to token constraints) preferredModels: - gpt5 # First choice for ≤400K tokens - gemini25pro # Second choice for ≤1M tokens - gpt41 # Fallback for ≤1M tokens # Default settings defaults: # Maximum tokens to use for model output maxOutputTokens: 32768 # Temperature for model responses temperature: 0 # Enable model fallback on API errors enableFallback: true # Default model per provider (used when no specific model is provided) providerModels: openai: gpt5 gemini: gemini25pro anthropic: opus41

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/jalehman/mcp-sage'

If you have feedback or need assistance with the MCP directory API, please join our Discord server