Ecosystem
ONNX
A portable model format that runs outside Python, including in browsers through WebGPU and in on-device runtimes.
- Category
- Ecosystem
- Also known as
- —
- Related terms
- 3
- Directory entries
- 1
- Docs
- —
- Added
- 2026-10-10
Definition
Exports put a decision model where the Python stack cannot go. Julia 1's WebGPU build loads an ONNX graph, 551 MB of external weights, and a Rust WASM tokenizer in the browser, and the onnx-community organization hosts Laya and d1 exports for runtimes beyond PyTorch.
Portability has a price to verify: browser parity is output agreement on a sample, 100 of 100 Julia predictions with a 0.00225 maximum logit difference, not a rerun of the accuracy suite. Check the artifact you actually deploy.
Tagsopen-modelsbrowser
Related terms
Definitions that connect to this one.
QuantizationStoring model weights at lower precision, 8-bit or 4-bit, to cut size and memory, at some risk to fine-grained probabilities.EcosystemOpen weightsModel checkpoints published for anyone to download, run, and fine-tune, as opposed to a hosted closed API like Jev.EcosystemLocal decision serverA process on your machine that answers typed questions from local weights, so the state does not leave the host.Ecosystem
From the Jev directory
A browser WebGPU export of Julia 1: the same 551 MB decision checkpoint, a Rust WASM tokenizer, and a JavaScript API that answers typed questions client-side with no server or key.
Jev Models & Reimplementations#community#open-models#local
Community
From the Jev community
Posts from builders shipping with Jev right now.