Test-network preview

Models

Status: test-network preview. Prices are suggestions; each provider sets its own.

The curated catalog below makes the common case one click in the macOS app. The network is model-agnostic: an operator can also install other weights that fit the machine, and the id column is the value of the OpenAI model field.

idNameParametersQuantizationMemoryDiskLicenseInput credits / 1MOutput credits / 1MNotes
llama-3.2-3b-instructLlama 3.2 3B Instruct3B4-bit4 GB1.8 GBLlama 3.2 Community License0.020.04
qwen3-4bQwen3 4B4B4-bit5 GB2.3 GBApache-2.00.030.06
llama-3.1-8b-instructLlama 3.1 8B Instruct8B4-bit7 GB4.5 GBLlama 3.1 Community License0.050.08
qwen3-8bQwen3 8B8B4-bit7 GB4.6 GBApache-2.00.050.1
qwen3-14bQwen3 14B14B4-bit11 GB8.3 GBApache-2.00.080.2
gemma-3-27b-itGemma 3 27B27B4-bit20 GB16 GBGemma Terms of Use0.10.25
qwen3-30b-a3bQwen3 30B A3B (MoE)30B total / 3B active4-bit21 GB17 GBApache-2.00.080.25
qwen3-32bQwen3 32B32B4-bit23 GB18 GBApache-2.00.150.4
llama-3.3-70b-instructLlama 3.3 70B Instruct70B4-bit46 GB40 GBLlama 3.3 Community License0.30.6
qwen3-235b-a22bQwen3 235B A22B (MoE)235B total / 22B active4-bit145 GB132 GBApache-2.00.41.2
muse-spark-openMuse Spark Open816B total / 42B activeNVFP4560 GB493 GBMuse Model License1.254.25cluster only

Columns

Licenses

License terms travel with the model. An operator who offers a model to the network is serving it commercially and takes on that license's obligations, such as attribution ("Built with Llama"), acceptable-use policies, and size or revenue thresholds that need a separate license. Those obligations are the operator's responsibility, not the network's. The app shows each license before you offer the model; read it in full at the linked page.

Every receipt names the exact weights the provider ran by their sha256 hash, so a buyer can tell which build of a model served the request.

This page as Markdown: models.md. Every Compute page for agents: llms.txt.