Modal

Code on Modal reaches other systems through what the customer writes: OIDC tokens authenticate functions to external clouds, cloud buckets mount as filesystems, secrets hold credentials, Slack notifications report on apps (beta), and examples connect Discord, Google Sheets and Tailscale. No catalog of connectors, OAuth actions or tool gateway of Modal's own is documented; the tools are the customer's code.

Sourcemodal.com/llms.txtread 2026-09-22

Retries are first class on Modal: a Retries policy can sit on any function, alongside job queues, fan out with map and spawn, batch and dynamic batching, and gang scheduled clusters, so a pipeline that mixes model calls with deterministic steps can be built on it. Sequencing and branching live in the customer's Python, and no multistep workflow primitive that records completed steps and resumes after failure is documented.

Sourcemodal.com/docs/guide/retriesread 2026-09-22

Retrieval systems the customer builds can run on Modal (its examples embed documents with TEI and run a RAG chatbot over PDFs), but Modal does not maintain a retrieval structure over the customer's content that an agent queries. Volumes, Dicts and mounted buckets are storage, not a Modal index a query returns, and nothing documented lets the product itself index, refresh or permission sources.

Sourcemodal.com/llms.txtread 2026-09-22

No approval step, consent checkpoint or escalation to a person is documented for work running on Modal. Restricted Functions come closest: they stop untrusted code from touching Modal resources, calling other functions or reaching Modal's APIs. Together with sandbox networking controls, they are isolation the developer configures, not a point where a person reviews an agent's action.

Sourcemodal.com/docs/guide/restricted-accessread 2026-09-22

Modal states a SOC 2 Type II audit, with the report available through its Security Portal at trust.modal.com, HIPAA support, and external penetration testing. Identity runs through SSO with Okta, Microsoft Entra or custom SAML, with SCIM provisioning (beta). Access is managed with role based access control, user groups and service users, and an append only audit log records sensitive actions on Enterprise. Secrets, Restricted Functions that cannot touch Modal resources, and sandbox networking controls round out the controls.

Sourcemodal.com/docs/guide/securityread 2026-09-22

The workload itself is well recorded: real time logs and metrics per function and container, GPU metrics, function stats, an append only audit log of sensitive workspace actions on Enterprise, and export of audit logs, function logs and container metrics to any OpenTelemetry provider or Datadog. Audit stays separate from runtime. Prompts, tool calls, retrieved knowledge and outputs can be inspected step by step only as far as the customer's own code logs them; no tracing of each call an agent makes is documented.

Sourcemodal.com/docs/guide/otel-integrationread 2026-09-22

State carries across runs: sandbox snapshots save and restore a sandbox's filesystem and memory, memory snapshots speed cold starts, volumes persist files, and Dicts and Queues hold data shared across function calls. The agent's code reads that state to carry on. Volumes and Dicts are storage the code reads, though, and restored process state belongs to the machine; the buyer can resume or delete it but not review or edit it as memory. No memory layer with a stated scope and lifetime is documented.

Sourcemodal.com/docs/guide/sandbox-snapshotsread 2026-09-22

Workloads can be pinned to a region. Functions and Sandboxes take a region argument for the container region, dedicated Endpoints set both container and routing regions, and a data residency guide sets out per service what can be pinned and when customer data may leave the region. Region selection carries a stated price multiplier (1.15x broad, 1.75x narrow). Modal is a managed cloud with no self hosted option, and shared Endpoints do not take a region.

Sourcemodal.com/docs/guide/data-residencyread 2026-09-22

Ready made starting points come as a gallery of runnable examples and a Library of models deployable as Shared or Dedicated Endpoints. The agent examples include Cursor cloud agents, a background coding agent with OpenCode, a coding platform, a Claude Agent SDK Slack bot, a LangGraph agent in a GPU sandbox, a computer use agent over VNC and parallel evals. These are examples and deployable models a developer builds from, not a browsable catalog of agents a buyer adopts.

Sourcemodal.com/llms.txtread 2026-09-22

Work starts without a person. Scheduled Functions run on a cron expression or a fixed period, web functions and endpoints boot a container from zero when a request arrives, and job queues hand work to functions as it lands.

Sourcemodal.com/docs/guide/cronread 2026-09-22

Model choice is the customer's throughout: Modal serves models from its Library on Shared Endpoints billed per token or Dedicated Endpoints with the customer's capacity and region, and the customer can deploy any open model on GPUs with vLLM or SGLang, as its examples show for DeepSeek, Nemotron and others.

Sourcemodal.com/docs/guide/shared-endpointsread 2026-09-22

Official SDKs in Python, JavaScript/TypeScript and Go (the latter two in beta) let outside code call Modal, alongside a full CLI that includes a skills command for coding agents. Web functions and endpoints expose the customer's code as HTTP APIs, and a gRPC API sits behind the open source client.

Sourcemodal.com/llms.txtread 2026-09-22

Evaluations run on Modal, but scoring them is left to other tools. Its example runs massively parallel evals with Harbor in Modal Sandboxes, and its reinforcement learning examples use verl and TRL, with the datasets, graders and scoring belonging to those frameworks. No fixtures, scoring, quality gates or comparison of agent runs are documented as Modal features.

Sourcemodal.com/llms.txtread 2026-09-22

Computer use agents can run on Modal's machines. Its example runs a Browser Use agent driving Chromium inside a Modal Sandbox, watched over VNC, with the model served from a Modal Endpoint, and VM Sandboxes are in beta. Control of the browser comes from Browser Use, a separate library, and no Modal API for screenshots, clicks or typing is documented: Modal supplies the environment, not the control.

Sourcemodal.com/docs/examples/computer_use_vncread 2026-09-22