Models / OpenJev

OpenJev

edge
HuggingFace → Find for my hardware

Open reproduction of the Jev interface, built two days after launch. Fine-tuned Qwen3 checkpoints (0.6B up) scored against fixed options with calibrated confidence. Runs fully client-side in the browser on WebGPU - works on a phone, no API or cloud. Accuracy ladder 0.407 to 0.845 against Jev’s published 0.883 on the shared subset.

AI-generated content marks

The provider reports that this model does not add embedded watermarks or provenance metadata to generated output.

general chat
Parameters
600M
Context
8k
License
mit
Developer
Undisclosed
Released
Sep 2026

Save your hardware and every model page answers the real question: will it run on your machine, and how fast?

Join free - save your rig →

Run it locally

Per-quant memory needs and a static "can you run it?" reference - no rig entry required

Q4_K_M
0.4GB 0.6GB min 1.0GB rec
Balanced - the usual local sweet spot
Q8_0
0.7GB 1.0GB min 1.6GB rec
Near-lossless
BF16
1.2GB 1.6GB min 2.4GB rec
Full quality, largest

The reference hardware

Schematic of the 4x H100 80GB (320GB) reference rig - 320GB VRAM, 13400 GB/s aggregate bandwidth
Schematic of the NVIDIA DGX Station 748GB reference rig - 748GB unified memory, 8000 GB/s aggregate bandwidth
Schematic of the 8x RTX 3090 rack (192GB) reference rig - 192GB VRAM, 7489 GB/s aggregate bandwidth
Schematic of the 4x RTX 5090 (128GB) reference rig - 128GB VRAM, 7168 GB/s aggregate bandwidth
Schematic of the AMD Instinct MI300X (192GB) reference rig - 192GB VRAM, 5324 GB/s aggregate bandwidth
Schematic of the 4x RTX 4090 (96GB) reference rig - 96GB VRAM, 4032 GB/s aggregate bandwidth
Schematic of the 2x RTX 5090 (64GB) reference rig - 64GB VRAM, 3584 GB/s aggregate bandwidth
Schematic of the 2x RTX 3090 (48GB) reference rig - 48GB VRAM, 1872 GB/s aggregate bandwidth
Schematic of the Single RTX 5090 (32GB) reference rig - 32GB VRAM, 1792 GB/s aggregate bandwidth
Schematic of the RTX PRO 6000 Blackwell (96GB) reference rig - 96GB VRAM, 1792 GB/s aggregate bandwidth
Schematic of the Mac Studio M4 Ultra 192GB reference rig - 192GB unified memory, 1092 GB/s aggregate bandwidth
Schematic of the Mac Studio M4 Ultra 512GB reference rig - 512GB unified memory, 1092 GB/s aggregate bandwidth
Schematic of the Single RTX 4090 (24GB) reference rig - 24GB VRAM, 1008 GB/s aggregate bandwidth
Schematic of the MacBook Pro M5 Max 128GB reference rig - 128GB unified memory, 614 GB/s aggregate bandwidth
Schematic of the Single GTX 1080 Ti (11GB) reference rig - 11GB VRAM, 484 GB/s aggregate bandwidth
Schematic of the Dual EPYC 9004 + 768GB DDR5-4800 reference rig - 768GB unified memory, 460 GB/s aggregate bandwidth
Schematic of the DGX Spark 128GB unified reference rig - 128GB unified memory, 273 GB/s aggregate bandwidth
Schematic of the Ryzen AI Max+ 395 128GB reference rig - 128GB unified memory, 256 GB/s aggregate bandwidth
Schematic of the Jetson AGX Orin 64GB reference rig - 64GB unified memory, 204 GB/s aggregate bandwidth
Schematic of the Epyc + 512GB DDR4-3200 + 2x RTX 3090 reference rig - 560GB unified memory, 204 GB/s aggregate bandwidth
Schematic of the Epyc + 512GB DDR4-2400 + 2x RTX 3090 reference rig - 560GB unified memory, 153 GB/s aggregate bandwidth
Schematic of the NVIDIA Jetson Orin NX 16GB reference rig - 16GB unified memory, 102 GB/s aggregate bandwidth

22 reference configs, drawn in-house. Scroll for more.

Can you run it? - reference rigs

Rig Q4_K_M Q8_0 BF16
4x H100 80GB (320GB) fast 19394.7t/s fast 11338.5t/s fast 6141.7t/s
NVIDIA DGX Station 748GB fast 11579.0t/s fast 6769.2t/s fast 3666.7t/s
8x RTX 3090 rack (192GB) fast 10840.2t/s fast 6337.4t/s fast 3432.7t/s
4x RTX 5090 (128GB) fast 10374.7t/s fast 6065.2t/s fast 3285.3t/s
AMD Instinct MI300X (192GB) fast 7707.0t/s fast 4505.6t/s fast 2440.5t/s
4x RTX 4090 (96GB) fast 5835.8t/s fast 3411.7t/s fast 1848.0t/s
2x RTX 5090 (64GB) fast 5187.4t/s fast 3032.6t/s fast 1642.7t/s
2x RTX 3090 (48GB) fast 2710.1t/s fast 1584.3t/s fast 858.2t/s
Single RTX 5090 (32GB) fast 2593.7t/s fast 1516.3t/s fast 821.3t/s
RTX PRO 6000 Blackwell (96GB) fast 2593.7t/s fast 1516.3t/s fast 821.3t/s
Mac Studio M4 Ultra 192GB fast 1724.2t/s fast 1008.0t/s fast 546.0t/s
Mac Studio M4 Ultra 512GB fast 1724.2t/s fast 1008.0t/s fast 546.0t/s
Single RTX 4090 (24GB) fast 1459.0t/s fast 852.9t/s fast 462.0t/s
MacBook Pro M5 Max 128GB fast 969.5t/s fast 566.8t/s fast 307.0t/s
Single GTX 1080 Ti (11GB) fast 700.5t/s fast 409.5t/s fast 221.8t/s
Dual EPYC 9004 + 768GB DDR5-4800 fast 667.0t/s fast 389.9t/s fast 211.2t/s
DGX Spark 128GB unified fast 395.1t/s fast 231.0t/s fast 125.1t/s
Ryzen AI Max+ 395 128GB fast 370.5t/s fast 216.6t/s fast 117.3t/s
Jetson AGX Orin 64GB fast 296.4t/s fast 173.3t/s fast 93.9t/s
Epyc + 512GB DDR4-3200 + 2x RTX 3090 fast 296.4t/s fast 173.3t/s fast 93.9t/s
Epyc + 512GB DDR4-2400 + 2x RTX 3090 fast 222.3t/s fast 130.0t/s fast 70.4t/s
NVIDIA Jetson Orin NX 16GB fast 148.2t/s fast 86.7t/s fast 46.9t/s

Fit tiers use the same will-it-run logic as the rig finder. For comfortable fits, the badge reflects decode speed: fast >=20 t/s, ok 8-20 t/s, slow <8 t/s. t/s is a bandwidth estimate, not a measured benchmark.

Download options

Q4_K_M -4% vs fp16
0.4GB dl 0.6GB min 1.0GB rec
REC RAM vs largest quant
GGUF on HF →
Q8_0 -1% vs fp16
0.7GB dl 1.0GB min 1.6GB rec
REC RAM vs largest quant
GGUF on HF →
BF16
1.2GB dl 1.6GB min 2.4GB rec
REC RAM vs largest quant
GGUF on HF →

Or run it in the cloud

No per-token API provider pricing tracked for OpenJev yet. For flagship list prices, see the calculator.

PRICE HISTORY

Inference cost over time

Data accumulates from the first daily sync - longer ranges populate over time. Prices come from OpenRouter snapshots, not a historical API.

Loading price history...