AgentKernel AI model data

Measured conclusion: Deep-tested. L2 outcome: tools_ok.

Live prices, rankings, licences and privacy of AI models: best model per task, costs, alternatives.

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "AgentKernel AI model data": {
      "url": "https://www.agentkernel.si/mcp",
      "transport": "streamable-http"
    }
  }
}

Measured tools (6)

• best_for_task read-only

  Ranks AI models for one task by performance only (price shown, never used for the rank). Tasks cover text, coding, agents, websites/UI/games/data viz, image generation and editing, video from text or image, video editing, text to speech, transcription, music and embeddings.

  Parameters:

  • task (required) — string — Task slug
  • limit — integer
  • license — string — open = open weights only, closed = proprietary only

  Hints: read-only yes · destructive not declared · idempotent yes · open-world no

• recommend_models read-only

  Returns the best-fitting language models for a use case and monthly API budget using the published AgentKernel Fit Score (performance, cost, context, features), with optional privacy and licence requirements.

  Parameters:

  • use_case (required) — string
  • privacy — string — zdr = zero data retention, eu = zero retention in an EU region
  • priority — string
  • include_free — boolean
  • input_tokens — integer — Input tokens per month
  • needs_vision — boolean
  • output_tokens — integer — Output tokens per month
  • open_weights_only — boolean
  • monthly_budget_usd — number

  Hints: read-only yes · destructive not declared · idempotent yes · open-world no

• get_model read-only

  Price, context window, performance/coding/agentic index with rank, licence, privacy, cost for a usage volume and cheaper or open alternatives for one language model.

  Parameters:

  • model (required) — string — Model id like "anthropic/claude-opus-5.5" or a name like "Claude Opus 5.5"
  • input_tokens — integer
  • output_tokens — integer

  Hints: read-only yes · destructive not declared · idempotent yes · open-world no

• find_models read-only

  Searches and filters the catalog of language models by name, provider, licence and privacy, sorted by performance, coding, agentic index, price or context.

  Parameters:

  • sort — string
  • limit — integer
  • query — string — Part of a name, id or provider
  • license — string
  • privacy — string

  Hints: read-only yes · destructive not declared · idempotent yes · open-world no

• estimate_cost read-only

  Monthly API cost at base rates for up to five models and a token volume, with the volume translated into words and pages.

  Parameters:

  • models (required) — array
  • input_tokens — integer
  • output_tokens — integer

  Hints: read-only yes · destructive not declared · idempotent yes · open-world no

• cheaper_alternative read-only

  For a model you use today, finds a cheaper model at most five performance-index points lower for the same usage.

  Parameters:

  • model (required) — string
  • input_tokens — integer
  • output_tokens — integer

  Hints: read-only yes · destructive not declared · idempotent yes · open-world no

Publisher website

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.