Skip to main content
Glama

chat_gemini_carbon

Запрос к Gemini 4 Carbon (medium) через вашу подписку Google AI Pro.

    Средняя линейка Gemini 4: дешевле и быстрее Hydrogen при похожем качестве
    на коротких задачах. Как и остальные Gemini 4, требует аккаунт, которому
    Carbon открыт: 404 от Google значит, что у подключённого аккаунта доступа
    нет — смотрите list_available_llm и переключайтесь через account_email.
    Варианты `gemini-4-carbon-low` / `-high` доступны по точному id через
    OpenAI-совместимый `/v1/chat/completions`.

    Args:
        messages: список {"role": "user|assistant", "content": "..."}.
        system: системный промпт (опционально).
        max_tokens: максимум токенов в ответе (1..65536); размышления входят в лимит.
        thinking_budget: 0 — выключить, -1 — динамически, N — верхняя граница.
        thinking_level: "low" | "medium" | "high" (приоритетнее thinking_budget).
        account_email: форсировать конкретный Google-аккаунт из пула.
        account_tag: форсировать аккаунты с указанным тэгом.

    Returns:
        {"text": str, "usage": {"input_tokens": int, "output_tokens": int}}.

    Raises:
        MCPError: ключ невалиден, Google не подключён, модели нет на аккаунте или квота исчерпана.
    

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "messages": "<messages>"
    },
    "name": "chat_gemini_carbon"
  }
}

Input Schema

NameRequiredDescriptionDefault
systemNo
messagesYes
max_tokensNo
account_tagNo
account_emailNo
thinking_levelNo
thinking_budgetNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Added

TDQS

A4.7/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already cover safety shape (readOnlyHint=false, destructiveHint=false, openWorldHint=true), but the description adds real behavioral context: access is gated per Google account, a Google 404 specifically means the connected account lacks access, thinking tokens count against max_tokens, and MCPError is raised on invalid key, disconnected Google, missing model, or exhausted quota. It stops short of describing pooling/failover behavior when an account is exhausted.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Purpose, tier positioning, and the access caveat are front-loaded before the Args/Returns/Raises blocks, so scannability is good and nearly every sentence carries usable information. A little redundancy exists around the model-variant note and the access explanation, but nothing is padding.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 7-parameter, quota-consuming, account-gated model call with no output schema, the definition covers invocation inputs, return shape ({text, usage}), and the failure modes an agent must handle. Nothing an agent needs in order to call it correctly or recover from an error appears to be missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, so the description carries the full load, and it does: messages shape, optional system prompt, max_tokens range (1..65536) with the note that reasoning tokens consume it, thinking_budget's 0/-1/N semantics, and thinking_level taking precedence over thinking_budget. account_email and account_tag are explained as forcing a specific account or tag from the pool, which no structured field conveys.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description names a specific verb and resource (a request to Gemini 4 Carbon in medium mode via the user's Google AI Pro subscription) and explicitly positions it against the Hydrogen sibling and the low/high id variants. An agent can distinguish it from chat_gemini_hydrogen and chat_gemini_flash without opening any schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It states when this tier is the right pick (cheaper/faster than Hydrogen while matching it on short tasks) and gives an explicit recovery path for the 404 case: check list_available_llm and switch accounts via account_email. It also routes to the low/high variants by exact model id, which is genuine when-to-use guidance rather than restatement.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.