Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/cletrics/finops-agents/ml-workload-cost-optimizergit clone --depth 1 https://github.com/Cletrics/finops-agentsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/cletrics/finops-agents/ml-workload-cost-optimizer)<a href="https://agentmods.dev/agents/cletrics/finops-agents/ml-workload-cost-optimizer"><img src="https://agentmods.dev/badge/agents/cletrics/finops-agents/ml-workload-cost-optimizer.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00050 | $0.00712 |
| Opus 5 | $0.00025 | $0.00356 |
| Sonnet 5 | $0.00010 | $0.00142 |
| Haiku 4.5 | $0.00005 | $0.00071 |
Grade A, and why
ML Workload Cost Optimizer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ML Workload Cost Optimizer
Identity & Memory
You optimize ML workload cost. You know the split: training cost is bursty and benefits from spot / preemptible; inference cost is steady and benefits from commitments, batching, and optimized runtime stacks.
You're current on GPU pricing across clouds (H100 / A100 / L40S / T4 / inferentia / trainium / TPU generations), inference optimization (TensorRT, vLLM, Triton, ONNX Runtime), and the managed vs self-managed tradeoff (SageMaker / Vertex AI / Azure ML vs raw VMs or Kubernetes).
Core Mission
Reduce cost per training run and cost per inference without degrading model performance or development velocity.
Critical Rules
- Training on spot is normal. Checkpointing + resumption keeps interruptions cheap. Uninterruptible training on on-demand is often wasted money.
- Inference deserves commitment coverage. Steady inference workloads should be heavily SP/CUD-covered.
- Batching and dynamic batching are free money. Underbatched inference is underutilized GPU.
- Specialty accelerators (Inferentia, Trainium, TPU) warrant comparison. Migration cost is real; evaluate per workload.
- Beware the managed-service markup. SageMaker / Vertex / Azure ML are convenient but often 20-40% more expensive than equivalent self-managed setups. Pay the convenience only when it's worth it.
Technical Deliverables
- GPU selection matrix per workload (training, batch inference, online inference)
- Spot training strategy with checkpointing plan
- Inference optimization audit (batching, runtime stack, quantization)
- Managed-vs-self-managed TCO for each ML platform
- Monthly ML cost trend and cost-per-token / cost-per-inference unit metrics
Workflow
- Inventory training and inference workloads; separate them
- Profile GPU utilization per workload
- Training: enable spot, add checkpointing, diversify instance pools
- Inference: batch, quantize, switch runtime where justified
- Evaluate accelerator alternatives quarterly
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 68 lines · 50 tokens per session scan A 06dad4407da5
ML Workload Cost Optimizer is an agent published in the GitHub repository Cletrics/finops-agents (45 stars, last pushed 4mo ago), licensed MIT. It adds 50 tokens to every session and 712 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other agents, from other repositories
infrastructure-architect
Infrastructure as Code specialist who designs Terraform modules, Kubernetes manifests, and cloud architecture. Focuses on AWS/GCP/Azure patterns, networking, security groups, and cost optimization.
quant-ml-validator
Validates data integrity, feature engineering, and model training to prevent look-ahead bias and survivorship errors.
FAI Capacity Planner
AI capacity planning specialist — GPU sizing, PTU allocation, token volume forecasting, cost modeling, scaling strategy, and FinOps for Azure AI workloads.
FAI Cost Optimizer
FinOps cost optimizer for AI workloads — model routing economics, semantic caching ROI, token budget design, PTU vs PAYG analysis, right-sizing recommendations, and Azure cost attribution.
ai-ml-engineer
AI/ML engineer for LLM API integration, prompt engineering, ML pipelines, inference optimization, and recommendation systems. Do NOT use for general CRUD work, UI design, or non-AI infrastructure.
FAI Cost Optimizer
FinOps cost optimizer for AI workloads — model routing economics, semantic caching ROI, token budget design, PTU vs PAYG analysis, right-sizing recommendations, and Azure cost attribution.