Skip to documentation

Support and compatibility

A small measured matrix is more useful than a long compatibility promise.

The first public beta is being qualified for one local Apple Silicon workflow. This page separates implemented requirements, directly measured evidence, and work that must still pass before a public download exists.

Status: release-candidate contract; no public app download

Reviewed:

Current platform boundary

Studio is built for Apple Silicon and a minimum deployment target of macOS 14. The complete native workflow has so far been measured on an Apple M4 Max with 128 GB unified memory. Representative lower-memory Macs remain in the clean-device matrix, so an exact public minimum is not yet approved.

Tensor Cortex Studio platform and resource boundary
AreaFirst-beta boundaryEvidence
ProcessorApple Silicon, arm64 onlyQualified on Apple M4 Max
Operating systemmacOS 14 or newerDeployment target implemented; full-workflow evidence currently on macOS 26.x
Unified memoryPublic minimum pendingDo not infer a minimum from one high-memory qualification host
Local storageAbout 755 MB app, 2.3 GB model, at least 5 GB free before downloadMeasured app/model size and tested disk preflight
InternetFirst model download and optional update checks onlyAnonymous model request and updater endpoint captured
Developer toolsNot required by the distributed appRelocatable Python/MLX engine passed; notarized artifact pending

The one first-beta workflow candidate

  • Model:mlx-community/Qwen3-4B-Instruct-2507-4bit at pinned revision50d427756c6b1b2fe0c0a10f67fbda1fc8e82c1b; model-card license metadata is Apache-2.0.
  • Data: UTF-8 train.jsonl, valid.jsonl, and test.jsonl, with one object per line and string prompt and answer fields.
  • Method: supervised LoRA/QLoRA through the bundled MLX runtime, with bounded Guided or Expert parameters.
  • Evidence: dataset health, frozen baseline, candidate evaluation against the same test split, run state, source revisions, and artifact hashes.
  • Export: a content-addressed LoRA adapter package with experiment, evaluation, and provenance evidence—not a fused model.

A project may remain open beside other projects, but only one operation runs inside each project. Concurrent training across projects has no public support guarantee and depends on available unified memory.

Explicitly not supported in the first beta

  • Intel Macs, Windows, Linux, iPhone, iPad, or virtualized macOS.
  • Generic document extraction, OCR, image/audio/video training, general chat, or agent workflows.
  • Fused full-model, GGUF, Ollama, Core ML, or Hugging Face publishing export.
  • Accounts, remote project storage, managed cloud compute, Multi-Mac execution, BYOC, billing, team roles, or sharing.
  • Product analytics, automatic crash reports, background diagnostics, or a diagnostic upload service.
  • A guarantee that fine-tuning improves quality or that an adapter works outside its recorded base-model/runtime boundary.

Start with the first incomplete stage

Safe actions for common Tensor Cortex Studio problems
ProblemCheckSafe action
Model will not downloadInternet access and at least 5 GB freeRetry from Baseline; verified existing caches are preserved
Baseline is unavailableData and model readinessReturn to the first incomplete stage shown in the journey
Training was interruptedRestored run stateUse Resume only when Studio offers it; otherwise start a new run
Mac is under memory pressureOther ML work and concurrent runsStop other workloads and run projects sequentially
Export is disabledCompleted run, candidate evaluation, and dataset revisionComplete the prerequisite named by Studio
Update check failsNetwork and updater settingKeep the installed version; never bypass signature checks

A support request should not become a data export

Include the Studio version/build, macOS version, chip and coarse memory tier, rounded free-disk range, workflow stage, exact visible error title/message and any displayed code, and concise reproduction steps. Crop or redact screenshots. Review every attachment.

Do not send by default: datasets, prompts, answers, model outputs, project/customer names, filenames or full paths, manifests, run/eval JSON, models, adapters, checkpoints, packages, credentials, environment variables, or unreviewed macOS diagnostic archives.

Studio has no automatic report or upload path. The local engine's developer-facing doctor command can create a redacted bundle from explicitly selected JSON/YAML, but it is not a public in-app send feature. The first beta sends no product telemetry or automatic crash report.

What remains before this matrix is operative

  1. Developer ID sign and notarize two application versions.
  2. Complete install, first launch, workflow, recovery, export, update, and uninstall on clean supported accounts.
  3. Test representative lower-memory Macs and publish the smallest passing unified-memory tier.
  4. Repeat the complete anonymous model download and network capture from the release account.
  5. Approve the release legal text and third-party notices.

Need help with a development build?

Send a minimal reproduction without private workload data. Security vulnerabilities use the dedicated security address.

hello@tensorcortex.com