Lattice
Skip to content

Multi-step, tool use, long-running

A ai agent stack.

AI agent · 100,000 req/mo. At the default case — 100k requests a month, no further constraints — drawn from the 112 tools already in the index. Each pick is a tool whose own skip-when is stated, because a stack built on tools it does not know how to overcommit to is a guess.

This is the default case, stated as such. Describe your case to change it — the recommendation recomputes, and anything it assumes comes back in the answer rather than staying silent.

The picks

  • Routing — Bifrost

    Workable

    A high-throughput OpenAI-compatible proxy with a small footprint.

    Watch out: You need the breadth of a full routing stack.

    Consider Cloudflare AI Gateway instead when edge caching and latency dominate, on Cloudflare already

  • Agents — LangChain

    Strong fit

    Tool use and state, in an ecosystem where the answers already exist.

    Watch out: You want a thin, legible core — this is a large surface.

    Consider Letta instead when agents whose memory is a first-class, inspectable and editable state

  • Workflows — DBOS

    Strong fit

    Durable workflows as ordinary Python functions and decorators.

    Watch out: You want a separate service to operate.

    Consider Fly Machines instead when hosting stateful containers that fits durable agent workers

  • Inference — KoboldCpp

    Workable

    Local generation with a UI tuned for interactive, long-form use.

    Watch out: You need headless serving at scale.

    Consider llama.cpp instead when the constraint is hardware, not throughput — no GPU, or an edge box

  • Guardrails — PyRIT

    Strong fit

    Generating and scoring attack prompts systematically.

    Watch out: You are not authorised to test the target.

    Consider garak instead when probing your own system for known vulnerability classes

  • Evals — Braintrust

    Strong fit

    Making evaluation, rather than tracing, the primary workflow.

    Watch out: You need open formats and self-hosting.

    Consider LangSmith instead when teams already deep in LangChain needing tracing and datasets

The costs and the risks

Estimated cost
$40–$238 per month — a heuristic band from query volume, not a vendor quote.
Confidence
87% — lower when constraints narrow the field.
Cost drivers
Braintrust

Biggest risk at the default case: An agent loop that cannot survive a deploy will eventually cause real damage. Durability and tool permissions are the controls, not the prompt.

A starting point, not a prescription. The recommendation at the default case is what the engine would tell a stranger with your workload; the recommendation at *your* case is what it tells you once it knows the constraints. The Stack Builder holds both — the builder link preselects this workload and starts with your volumes from here.