Your data stays on your machine
Prompts and decisions stay within the environment you choose.
Decision Models · Run locally
Choose an open-weight model, check the hardware it needs, and serve it through a Jev-compatible endpoint on a machine you control.
Prompts and decisions stay within the environment you choose.
Measure response time on the device and network you plan to use.
Call the same typed /v1/systemone shape used by hosted systems.
The local service binds to 127.0.0.1 unless you change it.
Published benchmark
Capability rank and score are read from the current catalogue. Model licences and local hardware requirements still apply.
Capability score 81.69
View hardware and model terms →Capability score 79.57
View hardware and model terms →Capability score 77.59
View hardware and model terms →Capability score 75.95
View hardware and model terms →Capability score 75.00
View hardware and model terms →Capability score 73.49
View hardware and model terms →Capability score 73.47
View hardware and model terms →Capability score 72.50
View hardware and model terms →Capability score 72.41
View hardware and model terms →Capability score 71.74
View hardware and model terms →Published benchmark
Capability rank and score are read from the current catalogue. Model licences and local hardware requirements still apply.
Capability score 68.01
View hardware and model terms →Capability score 64.81
View hardware and model terms →Capability score 62.15
View hardware and model terms →Capability score 58.73
View hardware and model terms →Capability score 57.62
View hardware and model terms →Capability score 55.62
View hardware and model terms →Capability score 54.25
View hardware and model terms →Capability score 53.34
View hardware and model terms →Capability score 51.46
View hardware and model terms →This model is not offered in the installer while its licensing review is pending.
Local model catalogue
18 models have an executable install recipe. Others are marked setup in preparation. Model pages show runtime, memory, and licence details.
27B text and image model.
Hardware estimate: NVIDIA RTX PRO 6000 Blackwell (96 GB) workstation · unverified26B text model.
Hardware estimate: NVIDIA RTX PRO 6000 Blackwell (96 GB) workstation · unverified12B text model.
Hardware estimate: Apple Mac mini M5 Pro (18-core CPU, 20-core GPU, 64 GB, 1 TB) · unverified31B text model.
Hardware estimate: NVIDIA RTX PRO 6000 Blackwell (96 GB) workstation · unverified31B text model.
Hardware estimate: NVIDIA RTX PRO 6000 Blackwell (96 GB) workstation · unverified4B text model.
Hardware estimate: NVIDIA GeForce RTX 5090 (32 GB) workstation · install tested4B text and image model.
Hardware estimate: Apple Mac mini M6 (16 GB, 256 GB) · unverified12B text and image model.
Hardware estimate: NVIDIA RTX PRO 6000 Blackwell (96 GB) workstation · unverified25.2B text and image model.
Hardware estimate: Apple Mac Studio M5 Ultra (30-core CPU, 64-core GPU, 96 GB, 1 TB) · unverified4B text and image model.
Hardware estimate: NVIDIA GeForce RTX 5090 (32 GB) workstation · unverified4B text and image model.
Hardware estimate: Apple Mac mini M5 Pro (15-core CPU, 16-core GPU, 24 GB, 512 GB) · unverified31B text model.
Hardware estimate: Apple Mac Studio M5 Ultra (30-core CPU, 64-core GPU, 96 GB, 1 TB) · unverified4B text and image model.
Hardware estimate: Apple Mac mini M5 Pro (18-core CPU, 20-core GPU, 64 GB, 1 TB) · unverified31B text model.
Hardware estimate: NVIDIA RTX PRO 6000 Blackwell (96 GB) workstation · unverified25.2B text and image model.
Hardware estimate: NVIDIA RTX PRO 6000 Blackwell (96 GB) workstation · unverified26.5B text and image model.
Hardware estimate: NVIDIA RTX PRO 6000 Blackwell (96 GB) workstation · unverified12B text model.
Hardware estimate: NVIDIA RTX PRO 6000 Blackwell (96 GB) workstation · unverified26B text model.
Hardware estimate: NVIDIA RTX PRO 6000 Blackwell (96 GB) workstation · unverifiedHow it works
Installer plans use the pinned model revision and the constraints recorded for each local variant.
Local by default
Installer releases are signed, model downloads are checked against their recorded hashes, and the local API listens on loopback unless configured otherwise. The installer adds no usage telemetry.
cosign verify-blob --bundle SHA256SUMS.sigstore.json --certificate-identity-regexp '^https://github.com/fstandhartinger/decisionmodels-local/\.github/workflows/release\.yml@refs/tags/v[0-9][^/]*$' --certificate-oidc-issuer https://token.actions.githubusercontent.com SHA256SUMSThe bootstrap checks the signed manifest automatically. For a manual check, download SHA256SUMS and SHA256SUMS.sigstore.json from the GitHub release, then run this command.
Licence, separately
Using the installer does not change a model's licence. Some weights limit commercial use or redistribution.
Questions
No. Each page lists the recorded memory requirements and supported platforms. Use the plan command to check the machine you intend to use.
The catalogue pins a model repository and revision. The installer uses that official source and checks recorded file hashes.
No. Model licences are separate from the installer licence; review the model's own terms before commercial use.
Yes. The remote command connects to a machine you control over SSH. The remote host still needs a supported runtime and compatible hardware.