# SAVRN Cloud > Public documentation and service plans for SAVRN Cloud, the proposed AI service platform from SAVRN Infrastructure Platform LLC. The public interactive console is a browser simulation. No listed Cloud model, deployment, balance or result establishes live service capacity. Use each page's availability statement. This index is a navigation aid. robots.txt controls crawler access; this file does not grant additional permissions. ## Pages - [Build research workflows you can understand and govern](https://savrn.com/cloud/index.md): Explore SAVRN Cloud’s prepared Qwen model candidates, guided research starters and project workflows. Model execution is Coming soon; samples run in your browser. HTML: https://savrn.com/cloud/ - [Compare prepared models for the work you need to do](https://savrn.com/cloud/models.md): Compare SAVRN Cloud’s prepared Qwen3.5-9B Q6_K and Qwen3.8-27B Q5_K_M configurations, proposed limits and qualification status. Connected execution is Coming soon. HTML: https://savrn.com/cloud/models - [Prepare a useful request, then inspect the sample record](https://savrn.com/cloud/playground.md): Prepare evidence review, structured extraction or research-code requests in SAVRN Cloud. Inspect intended Qwen candidates and clearly labeled browser sample receipts. HTML: https://savrn.com/cloud/playground - [Give research activity a clear project context](https://savrn.com/cloud/projects.md): Explore project organization, sample budgets and linked research activity in SAVRN Cloud. Hosted project access and enforced team permissions are coming soon. HTML: https://savrn.com/cloud/projects - [Match execution choices to the work you need to run](https://savrn.com/cloud/compute.md): Explore workload placement and compute configuration in SAVRN Cloud’s simulated console. Connected rented and owned infrastructure options are coming soon. HTML: https://savrn.com/cloud/compute - [Understand a deployment from creation to retirement](https://savrn.com/cloud/deployments.md): Explore deployment configuration, request draining, restart and deletion in SAVRN Cloud’s public preview. Connected deployment control is coming soon. HTML: https://savrn.com/cloud/deployments - [Make a task’s execution boundary explicit](https://savrn.com/cloud/sandboxes.md): Explore bounded task execution, sample limits and exported results in SAVRN Cloud. The preview runs built-in browser tasks; connected sandboxes are coming soon. HTML: https://savrn.com/cloud/sandboxes - [Evaluate the behavior your research task requires](https://savrn.com/cloud/evaluations.md): Explore task-specific baselines, case results and evaluation receipts in SAVRN Cloud. Public preview scores are synthetic; connected evaluations are coming soon. HTML: https://savrn.com/cloud/evaluations - [Turn task expectations into a reviewable definition](https://savrn.com/cloud/environments.md): Explore task definitions, versioned rubrics and pass criteria in SAVRN Cloud’s public preview. Connected evaluation and training environments are coming soon. HTML: https://savrn.com/cloud/environments - [Know which data a research workflow is allowed to use](https://savrn.com/cloud/datasets.md): Discover datasets in SAVRN’s existing catalog and explore planned Cloud workflows for provenance, rights, evaluation splits and controlled dataset revisions. HTML: https://savrn.com/cloud/datasets - [Treat model adaptation as a controlled experiment](https://savrn.com/cloud/training.md): Explore adapter recipes, dataset purpose and candidate handoff in SAVRN Cloud. Training runs are simulated; connected model adaptation is coming soon. HTML: https://savrn.com/cloud/training - [Keep a promising candidate separate from an active release](https://savrn.com/cloud/candidates.md): Explore candidate evaluation, reviewer approval, promotion and rollback in SAVRN Cloud. All releases in the public preview are browser-local simulations. HTML: https://savrn.com/cloud/candidates - [Define the work and the result you expect](https://savrn.com/cloud/work-orders.md): Explore scoped research Work Orders, approval, sample execution and deliverable acceptance in SAVRN Cloud. Connected research workflows are coming soon. HTML: https://savrn.com/cloud/work-orders - [Make authorization a specific, recorded decision](https://savrn.com/cloud/approvals.md): Explore how SAVRN Cloud separates approval authority, task scope and budget decisions from execution and deliverable acceptance. Connected approvals are coming soon. HTML: https://savrn.com/cloud/approvals - [Connect resource use to the work that produced it](https://savrn.com/cloud/usage.md): Explore project usage, reservations and sample Work Credit receipts in SAVRN Cloud. The public preview illustrates accounting without charges or payments. HTML: https://savrn.com/cloud/usage - [Scope access to the actions an application needs](https://savrn.com/cloud/keys.md): Explore the planned lifecycle of scoped project credentials in SAVRN Cloud. Preview keys are simulated values; live API access and SDKs are coming soon. HTML: https://savrn.com/cloud/keys - [Separate resource inventory from usable capacity](https://savrn.com/cloud/capacity.md): Explore how SAVRN Cloud distinguishes resource inventory, workload qualification and available capacity. All provider and owned-resource records are simulated. HTML: https://savrn.com/cloud/capacity - [Turn operational events into understandable actions](https://savrn.com/cloud/operations.md): Explore readiness checks, incident states and operational evidence in SAVRN Cloud’s public preview. Connected monitoring and service operations are coming soon. HTML: https://savrn.com/cloud/operations - [Keep the research workflow portable as compute changes](https://savrn.com/cloud/architecture.md): Explore SAVRN Cloud’s planned separation of research workflows, governance and execution, including a qualified path from rented capacity to owned infrastructure. HTML: https://savrn.com/cloud/architecture - [Understand what must be ready before a service launches](https://savrn.com/cloud/roadmap.md): Review SAVRN Cloud’s planned service progression and the distinction between a public preview, qualified capability and a connected customer release. HTML: https://savrn.com/cloud/roadmap - [Understand which settings are preferences and which need enforcement](https://savrn.com/cloud/settings.md): Explore workspace preferences, session export and reset in SAVRN Cloud. Preview retention and placement choices are local settings, not enforced service policies. HTML: https://savrn.com/cloud/settings - [Welcome to SAVRN Cloud](https://savrn.com/cloud/docs/start-welcome.md): Explore SAVRN Cloud’s prepared Qwen models, guided research starters and browser-local sample receipts. Connected execution and customer capacity are Coming soon. HTML: https://savrn.com/cloud/docs/start-welcome - [What is available](https://savrn.com/cloud/docs/start-service-status.md): Understand prepared Qwen candidates, ready fictional deployments and completed samples in SAVRN Cloud, where connected inference remains Coming soon. HTML: https://savrn.com/cloud/docs/start-service-status - [Models, offers and deployments](https://savrn.com/cloud/docs/concepts-models-offers-deployments.md): Follow catalog models, prepared Qwen artifacts, qualified configurations, offers and deployments in SAVRN Cloud, with candidate and sample-fixture identities kept distinct. HTML: https://savrn.com/cloud/docs/concepts-models-offers-deployments - [First research Work Order](https://savrn.com/cloud/docs/start-first-work-order.md): Create a sample SAVRN Cloud Work Order, define its deliverable and budget, approve its scope, then inspect and accept the simulated research artifact. HTML: https://savrn.com/cloud/docs/start-first-work-order - [First inference walkthrough](https://savrn.com/cloud/docs/start-first-api-request.md): Prepare evidence review, structured extraction or research-code assistance in SAVRN Cloud and inspect a labeled sample receipt, without calling or executing a model. HTML: https://savrn.com/cloud/docs/start-first-api-request - [Projects and access](https://savrn.com/cloud/docs/platform-projects-access.md): Organize sample work by project in SAVRN Cloud, track allowances and understand the distinction between browser context and enforced production access. HTML: https://savrn.com/cloud/docs/platform-projects-access - [Data and approval policy](https://savrn.com/cloud/docs/platform-data-policy.md): Review SAVRN Cloud's preview data boundaries and the distinct permissions needed for datasets, execution and model releases, using public or invented inputs. HTML: https://savrn.com/cloud/docs/platform-data-policy - [Work Credits and usage](https://savrn.com/cloud/docs/platform-work-credits.md): Understand sample Work Credits in SAVRN Cloud, including allowances, reservations and receipts, while separating supplier cost from future customer pricing. HTML: https://savrn.com/cloud/docs/platform-work-credits - [Errors, retries and support](https://savrn.com/cloud/docs/platform-errors-support.md): Diagnose SAVRN Cloud preview errors, preserve useful request evidence and understand why rejected, failed and uncertain work need different retry behavior. HTML: https://savrn.com/cloud/docs/platform-errors-support - [Model discovery methodology](https://savrn.com/cloud/docs/models-discovery-methodology.md): Compare prepared Qwen artifacts and canonical SAVRN Model Hub records, with intended tasks, proposed limits and qualification gaps instead of unsupported rankings. HTML: https://savrn.com/cloud/docs/models-discovery-methodology - [Inference overview](https://savrn.com/cloud/docs/inference-overview.md): Explore SAVRN-operated inference preparation with Qwen utility and coding candidates, guided research starters and browser-local samples. Execution is Coming soon. HTML: https://savrn.com/cloud/docs/inference-overview - [Offers, rates and limits](https://savrn.com/cloud/docs/inference-offers-pricing.md): Read proposed Qwen pilot limits, illustrative usage and sample Work Credits separately from qualified SAVRN Cloud offers and approved customer pricing. HTML: https://savrn.com/cloud/docs/inference-offers-pricing - [Streaming, tools and limits](https://savrn.com/cloud/docs/inference-streaming-tools-limits.md): Separate SAVRN Cloud starter instructions from qualified streaming, schema, tool and context support, including the two Qwen candidates’ proposed pilot bounds. HTML: https://savrn.com/cloud/docs/inference-streaming-tools-limits - [Deployment lifecycle](https://savrn.com/cloud/docs/inference-deployments.md): Follow SAVRN Cloud's simulated deployment lifecycle from configuration to readiness, draining, restart and deletion, with logs and resource identity retained. HTML: https://savrn.com/cloud/docs/inference-deployments - [Compute overview](https://savrn.com/cloud/docs/compute-overview.md): Explore SAVRN Cloud compute planning across hosted, rented and future owned placement, including configuration review, resource limits and qualification needs. HTML: https://savrn.com/cloud/docs/compute-overview - [Rented compute preview](https://savrn.com/cloud/docs/compute-rented-digitalocean.md): Plan rented GPU execution in SAVRN Cloud, using DigitalOcean as an illustrative infrastructure option and checking quota, cost, networking and cleanup. HTML: https://savrn.com/cloud/docs/compute-rented-digitalocean - [Storage, networking and cleanup](https://savrn.com/cloud/docs/compute-storage-network-cleanup.md): Review storage, network access and cleanup in SAVRN Cloud, separating retained artifacts, stopped tasks and deleted allocations from actual billing evidence. HTML: https://savrn.com/cloud/docs/compute-storage-network-cleanup - [Move to owned capacity](https://savrn.com/cloud/docs/compute-owned-transition.md): Understand SAVRN Cloud's planned move from rented to owned compute, preserving project history while requalifying model identity, limits, costs and rollback. HTML: https://savrn.com/cloud/docs/compute-owned-transition - [Sandboxes overview](https://savrn.com/cloud/docs/sandboxes-overview.md): Explore SAVRN Cloud sandbox concepts, including task lifetime, resource limits, artifacts and isolation evidence, through clearly labeled browser simulations. HTML: https://savrn.com/cloud/docs/sandboxes-overview - [First sandbox run](https://savrn.com/cloud/docs/sandboxes-first-run.md): Create a sample sandbox task in SAVRN Cloud, inspect its output and logs, stop a run and export the result without executing code or allocating real resources. HTML: https://savrn.com/cloud/docs/sandboxes-first-run - [Images, secrets and egress](https://savrn.com/cloud/docs/sandboxes-images-secrets-egress.md): Review image provenance, job-scoped secrets and outbound network boundaries for SAVRN Cloud sandboxes, distinguishing planned settings from tested enforcement. HTML: https://savrn.com/cloud/docs/sandboxes-images-secrets-egress - [Environments overview](https://savrn.com/cloud/docs/environments-overview.md): Learn how SAVRN Cloud environments combine tasks, datasets, tools and scoring, with versioned preview records that connect an experiment to its evaluation. HTML: https://savrn.com/cloud/docs/environments-overview - [Build and test an environment](https://savrn.com/cloud/docs/environments-build-test.md): Configure a sample SAVRN Cloud evaluation environment, review its rubric and limits, then inspect the evidence needed for repeatable tasks and scoring. HTML: https://savrn.com/cloud/docs/environments-build-test - [Private environment registry](https://savrn.com/cloud/docs/environments-private-registry.md): Understand private environment releases in SAVRN Cloud, including ownership, immutable versions, publication permissions and the limits of browser-local records. HTML: https://savrn.com/cloud/docs/environments-private-registry - [Evaluations overview](https://savrn.com/cloud/docs/evaluations-overview.md): Plan interpretable model evaluations in SAVRN Cloud using versioned tasks, per-case evidence and usage records, while keeping synthetic preview scores distinct. HTML: https://savrn.com/cloud/docs/evaluations-overview - [Run a first baseline](https://savrn.com/cloud/docs/evaluations-first-baseline.md): Run a simulated baseline evaluation in SAVRN Cloud, freeze the task context, inspect case results and export a labeled receipt for later candidate comparisons. HTML: https://savrn.com/cloud/docs/evaluations-first-baseline - [Compare evaluation results](https://savrn.com/cloud/docs/evaluations-compare-results.md): Compare SAVRN Cloud evaluation records using matched tasks, failure cases, quality, latency and cost, without treating simulated scores as measured model gains. HTML: https://savrn.com/cloud/docs/evaluations-compare-results - [Lab and learning methods](https://savrn.com/cloud/docs/lab-overview.md): Explore SAVRN Cloud's model improvement workflow, from prompt and evaluation changes to proposed adapter training, candidate review and controlled release. HTML: https://savrn.com/cloud/docs/lab-overview - [Dataset and experiment releases](https://savrn.com/cloud/docs/lab-datasets-releases.md): Trace datasets into SAVRN Cloud experiments with purpose, rights notes, versions and held-out splits, using synthetic examples rather than real research records. HTML: https://savrn.com/cloud/docs/lab-datasets-releases - [First adapter training run](https://savrn.com/cloud/docs/lab-first-adapter-run.md): Follow a simulated LoRA or QLoRA recipe in SAVRN Cloud, inspect training context and candidate records, and learn what real checkpoint qualification requires. HTML: https://savrn.com/cloud/docs/lab-first-adapter-run - [Promote, deploy and roll back](https://savrn.com/cloud/docs/lab-promote-deploy-rollback.md): Review a SAVRN Cloud candidate, evaluate it on held-out data, record approval and rehearse promotion or rollback while preserving deployment and review evidence. HTML: https://savrn.com/cloud/docs/lab-promote-deploy-rollback - [Agents, sessions and recovery](https://savrn.com/cloud/docs/agents-overview-recovery.md): Understand bounded agent workflows in SAVRN Cloud, separating session history, approval and retained effects from durable execution and verified recovery. HTML: https://savrn.com/cloud/docs/agents-overview-recovery - [Governed literature-review walkthrough](https://savrn.com/cloud/docs/guides-literature-review.md): Rehearse a governed literature review in SAVRN Cloud: define sources and deliverables, approve scope, inspect a sample evidence table and record acceptance. HTML: https://savrn.com/cloud/docs/guides-literature-review - [Build a specialist research workflow](https://savrn.com/cloud/docs/guides-specialist-agent.md): Design a narrow research workflow in SAVRN Cloud using datasets, task rubrics and baselines before considering training, tool autonomy or candidate release. HTML: https://savrn.com/cloud/docs/guides-specialist-agent - [SDK and CLI compatibility](https://savrn.com/cloud/docs/developers-sdk-cli.md): Understand SAVRN Cloud's SDK and CLI publication boundary, what preview exports can show and the compatibility evidence needed before a client is released. HTML: https://savrn.com/cloud/docs/developers-sdk-cli - [Public API publication](https://savrn.com/cloud/docs/reference-api.md): See what a SAVRN Cloud API release must document: access scope, supported operations, limits, errors and usage receipts, distinct from browser preview routes. HTML: https://savrn.com/cloud/docs/reference-api - [Schemas and service limits](https://savrn.com/cloud/docs/reference-schemas-limits.md): Interpret SAVRN Cloud sample records and learn how service schemas should define units, revisions, limits and unknown values without inventing measured data. HTML: https://savrn.com/cloud/docs/reference-schemas-limits - [Deployment and trust matrix](https://savrn.com/cloud/docs/trust-deployment-matrix.md): Compare rented and owned SAVRN-operated compute boundaries, including verified model identity, data location, allocated capacity and operating responsibility. HTML: https://savrn.com/cloud/docs/trust-deployment-matrix - [Changelog and operational status](https://savrn.com/cloud/docs/resources-changelog-status.md): Review SAVRN Cloud's initial preview scope, browser-state behavior and the distinction between documentation updates, simulated activity and live service status. HTML: https://savrn.com/cloud/docs/resources-changelog-status - [Benchmarks and research methods](https://savrn.com/cloud/docs/research-benchmarks-methodology.md): Plan comparable Qwen evaluations from evidence review, extraction and research-code starters, keeping publisher claims and simulated results separate from SAVRN measurements. HTML: https://savrn.com/cloud/docs/research-benchmarks-methodology ## Machine-readable references - [XML sitemap](https://savrn.com/cloud/sitemap.xml) - [JSON content manifest](https://savrn.com/cloud/catalog.json) - [Complete public text](https://savrn.com/cloud/llms-full.txt)