accueil / serveurs / IA & Mémoire / io.github.helphub369/model-ruler
io.github.helphub369/model-ruler
serveur MCP (Model Context Protocol) · modelruler-inference-calculators 1.0.4
Référencé dans l'annuaire. Seul le pentest comportemental fait foi, et il n'a pas encore été exécuté sur ce serveur.
Est-ce que io.github.helphub369/model-ruler est sûr ?
Au dernier audit MCP Hub, le serveur MCP « io.github.helphub369/model-ruler » est classé « Non encore testé ». Référencé dans l'annuaire. Seul le pentest comportemental fait foi, et il n'a pas encore été exécuté sur ce serveur. Le verdict est réévalué en continu ; consultez la fiche pour l'état à jour.
Méthode : MCP Hub analyse la description de chaque outil (couche 1) et, pour le verdict fort, exécute le serveur dans une micro-VM isolée avec leurres et réseau bloqué (couche 2). On publie des noms et des compteurs, jamais le contenu brut renvoyé par un outil.
Les 12 outils
- agent-loop-cost-calculator≈ 416 tokens
Use for LLM inference/reasoning cost per successful multi-step agent task, including failure overhead and context growth across turns. Use agent-workflow-cost-calculator for iPaaS/app-action/MCP orchestration fees. Use automation-cost-calculator for generic Zapier/Make/n8n workflow billing.
Paramètres, retour & exemple (9)
Paramètresmodel· string · optionnel
LLM model (default claude-sonnet-4-6)provider· string · optionnel
LLM provider (default anthropic)success_rate· number · optionnel
Task success rate 0-1 (default 0.7)steps_per_task· number · optionnel
Avg reasoning steps per task (default 5)tasks_per_month· number · requis
Tasks attempted per monthtool_calls_per_step· number · optionnel
Avg tool invocations per step (default 2)context_growth_factor· number · optionnel
Multiplier on input tokens as conversation grows (default 1.4)avg_tokens_in_per_turn· number · optionnel
Avg input tokens per turn (default 3000)avg_tokens_out_per_turn· number · optionnel
Avg output tokens per turn (default 400)
Exemple d'appel{"name": "agent-loop-cost-calculator", "arguments": {"tasks_per_month": 1}}Exemple de retour{"content": [{"type": "text", "text": "…"}]} - agent-workflow-cost-calculator≈ 308 tokens
Use for automation-platform/iPaaS cost of an AI agent workflow, including app-action fan-out and MCP tool-call accounting, explicitly excluding LLM token spend. Use automation-cost-calculator for generic non-agent workflow billing. Use agent-loop-cost-calculator for LLM inference/reasoning cost per successful multi-step agent task.
Paramètres, retour & exemple (4)
Paramètresapp_actions_per_run· number · optionnel
Downstream app actions per agent run (default 3)agent_runs_per_month· number · requis
Agent workflow runs per monthmake_modules_per_run· number · optionnel
Make modules per run; defaults to app actions + MCP callsmcp_tool_calls_per_run· number · optionnel
MCP tool calls per agent run (default 1)
Exemple d'appel{"name": "agent-workflow-cost-calculator", "arguments": {"agent_runs_per_month": 1}}Exemple de retour{"content": [{"type": "text", "text": "…"}]} - automation-cost-calculator≈ 299 tokens
Use for generic or non-agent recurring workflow platform billing across Zapier task billing, Make credit billing, and n8n execution billing. Use agent-workflow-cost-calculator instead when the workflow is explicitly an AI agent with app actions or MCP tool calls. Use agent-loop-cost-calculator instead when the question is about LLM inference/reasoning cost per successful agent task.
Paramètres, retour & exemple (3)
Paramètresmake_modules_per_run· number · optionnel
Make module actions per scenario run; defaults to billable_steps_per_runbillable_steps_per_run· number · optionnel
Billable actions/modules per run (default 5)workflow_runs_per_month· number · requis
Workflow/scenario runs per month
Exemple d'appel{"name": "automation-cost-calculator", "arguments": {"workflow_runs_per_month": 1}}Exemple de retour{"content": [{"type": "text", "text": "…"}]} - context-window-planner≈ 289 tokens
Use when a user needs to know whether a document plus prompt plus output fits within a model's context window, or wants a strategy recommendation (truncate/summarize/rag/chunk).
Paramètres, retour & exemple (5)
Paramètresdoc_tokens· number · requis
Primary document/content tokensstrategy_hint· string · optionnel · truncate | summarize | rag | chunk
Preferred strategy (optional)overhead_tokens· number · optionnel
System prompt + few-shot + history (default 2000)expected_out_tokens· number · optionnel
Reserved output budget (default 1000)model_context_window· number · requis
Target model context size
Exemple d'appel{"name": "context-window-planner", "arguments": {"doc_tokens": 1, "model_context_window": 1}}Exemple de retour{"content": [{"type": "text", "text": "…"}]} - eval-cost-calculator≈ 372 tokens
Use when a user needs to budget an LLM evaluation run. Given samples/models/trials, returns total cost, per-run cost, and parallel time estimate.
Paramètres, retour & exemple (10)
Paramètresmodel· string · optionnel
Eval model (default claude-sonnet-4-6)models· number · optionnel
Candidate models (default 1)samples· number · requis
Number of eval samplesprovider· string · optionnel
Eval provider (default anthropic)avg_tokens_in· number · optionnel
Avg input tokens per sample (default 2000)judge_enabled· boolean · optionnel
Enable LLM-as-judge second pass (default false)avg_tokens_out· number · optionnel
Avg output tokens per sample (default 500)judge_tokens_in· number · optionnel
Judge input tokens (default 1500)judge_tokens_out· number · optionnel
Judge output tokens (default 200)trials_per_sample· number · optionnel
Repeats per sample (default 1)
Exemple d'appel{"name": "eval-cost-calculator", "arguments": {"samples": 1}}Exemple de retour{"content": [{"type": "text", "text": "…"}]} - fine-tune-roi-calculator≈ 324 tokens
Use when a user is considering fine-tuning vs prompt engineering. Returns training cost, monthly inference savings, months-to-ROI, and breakeven volume.
Paramètres, retour & exemple (6)
Paramètrestrain_tokens· number · requis
Training tokens (dataset × epochs)train_cost_per_1m· number · optionnel
Training cost per 1M tokensprompt_reduction_pct· number · optionnel
Prompt size reduction % from eliminating few-shot (default 0)base_inference_cost_1m· number · optionnel
Baseline API output cost per 1Mmonthly_inference_tokens· number · requis
Expected monthly inference volume (output tokens)finetuned_inference_cost_1m· number · optionnel
Fine-tuned inference cost per 1M
Exemple d'appel{"name": "fine-tune-roi-calculator", "arguments": {"train_tokens": 1, "monthly_inference_tokens": 1}}Exemple de retour{"content": [{"type": "text", "text": "…"}]} - observability-cost-calculator≈ 259 tokens
Use when a user needs to budget LLM observability tooling. Returns monthly cost at given request volume with retention adjustment.
Paramètres, retour & exemple (4)
Paramètresprovider· string · requis · langsmith | langfuse | helicone | portkey | arize | braintrust
Observability provideravg_log_bytes· number · optionnel
Avg payload bytes per traced request (default 4096)retention_days· number · optionnel
Retention in days (default 30)requests_per_day· number · requis
Average daily LLM requests
Exemple d'appel{"name": "observability-cost-calculator", "arguments": {"provider": "langsmith", "requests_per_day": 1}}Exemple de retour{"content": [{"type": "text", "text": "…"}]} - provider-cost-calculator≈ 298 tokens
Use when a user asks what an LLM workload costs on a specific provider/model, or wants to compare cost across providers. Given tokens per call and call volume, returns monthly cost plus a tier comparison table.
Paramètres, retour & exemple (6)
Paramètresmodel· string · optionnel
Target model (e.g. claude-sonnet-4-6)provider· string · optionnel
Target provider (e.g. anthropic, openai, together)tokens_in· number · requis
Input tokens per calltokens_out· number · requis
Output tokens per callcalls_per_month· number · optionnel
Monthly call volumeinclude_comparison· boolean · optionnel
Include tier comparison table (default true)
Exemple d'appel{"name": "provider-cost-calculator", "arguments": {"tokens_in": 1, "tokens_out": 1}}Exemple de retour{"content": [{"type": "text", "text": "…"}]} - quantization-calculator≈ 310 tokens
Use when a user is planning to quantize an LLM to fit on smaller hardware. Given parameter count and precision transition, returns VRAM requirement, speedup estimate, and approximate quality delta.
Paramètres, retour & exemple (5)
Paramètresbatch_size· number · optionnel
Serving batch size (default 1)precision_to· string · optionnel · fp32 | fp16 | bf16 | fp8 | int8 | int6 | int5 | int4
Target precision (default int4)precision_from· string · optionnel · fp32 | fp16 | bf16 | fp8 | int8 | int6 | int5 | int4
Starting precision (default bf16)kv_cache_tokens· number · optionnel
Max KV cache tokens (default 8192)params_billions· number · requis
Model parameter count in billions
Exemple d'appel{"name": "quantization-calculator", "arguments": {"params_billions": 1}}Exemple de retour{"content": [{"type": "text", "text": "…"}]} - rag-pipeline-cost-calculator≈ 471 tokens
Use when a user needs end-to-end RAG cost estimation (embedding + vector store + generation). Returns monthly cost with breakdown and dominant-component identification.
Paramètres, retour & exemple (11)
Paramètresvector_store· string · optionnel · pinecone | weaviate | qdrant | chroma | turbopuffer
Vector storeanswer_tokens· number · optionnel
Avg answer tokens (default 400)corpus_tokens· number · requis
Indexed corpus size in tokensgenerator_model· string · optionnel
Generator model (default claude-sonnet-4-6)queries_per_day· number · requis
User query volume per dayquestion_tokens· number · optionnel
Avg question tokens (default 100)chunks_retrieved· number · optionnel
Chunks per query (default 5)chunk_size_tokens· number · optionnel
Avg chunk size (default 512)embedding_provider· string · optionnel · openai-3-small | openai-3-large | voyage-3 | voyage-3-lite | cohere-embed-v3 | google-text-embed-004
Embedding modelgenerator_provider· string · optionnel
Generator LLM provider (default anthropic)reindex_fraction_per_month· number · optionnel
Fraction of corpus re-embedded per month (default 0.1)
Exemple d'appel{"name": "rag-pipeline-cost-calculator", "arguments": {"corpus_tokens": 1, "queries_per_day": 1}}Exemple de retour{"content": [{"type": "text", "text": "…"}]} - self-host-breakeven-calculator≈ 299 tokens
Use when a user is deciding between API usage and self-hosted GPU inference at a given volume. Returns breakeven token volume, monthly cost comparison, and go/no-go recommendation.
Paramètres, retour & exemple (6)
Paramètresgpu_type· string · optionnel
GPU type (e.g. h100, a100-80gb)gpu_provider· string · optionnel
GPU provider (e.g. runpod, modal)monthly_tokens· number · requis
Monthly output token volumeutilization_pct· number · optionnel
Expected GPU utilization % (default 60)api_cost_per_1m_out· number · optionnel
Current API output cost per 1M tokensoperational_overhead_pct· number · optionnel
Ops overhead % on GPU cost (default 40)
Exemple d'appel{"name": "self-host-breakeven-calculator", "arguments": {"monthly_tokens": 1}}Exemple de retour{"content": [{"type": "text", "text": "…"}]} - token-counter≈ 244 tokens
Use when a user asks how many tokens a given text will consume, or needs to estimate prompt size before pricing a workload. Given text and tokenizer family, returns low/high token range and byte-level measurements.
Paramètres, retour & exemple (3)
Paramètrestext· string · requis
Text content to count tokens fortokenizer· string · optionnel · cl100k_base | o200k_base | claude | gemini | llama3 | mistral | default
Tokenizer family (default: default)expected_out_tokens· number · optionnel
Expected output token budget (optional)
Exemple d'appel{"name": "token-counter", "arguments": {"text": "texte"}}Exemple de retour{"content": [{"type": "text", "text": "…"}]}
C'est votre serveur ?
Revendiquez-le sur l'espace créateur (preuve GitHub, domaine ou npm) pour afficher le badge « propriétaire vérifié » et lancer un pentest comportemental à la demande afin d'obtenir le badge « Vérifié ».
Source
https://modelruler.dev/mcp?dist=model_everywhere_registry_native_official_registry