Jonna Matthiesen's picture

Jonna Matthiesen

JonnaMat

embedl

·

AI & ML interests

None yet

Recent Activity

new activity about 19 hours ago

embedl/Edge-Inference-Benchmarks:Add accuracy data

updated a Space about 20 hours ago

embedl/Edge-Inference-Benchmarks

new activity about 20 hours ago

embedl/Edge-Inference-Benchmarks:Add subgroups to sidebar

View all activity

Organizations

New activity in embedl/Edge-Inference-Benchmarks about 19 hours ago

Add accuracy data

#4 opened about 19 hours ago by

updated a Space about 20 hours ago

Edge Inference Benchmarks

On-Device benchmarks across devices and models.

New activity in embedl/Edge-Inference-Benchmarks about 20 hours ago

Add subgroups to sidebar

#3 opened about 22 hours ago by

replied to their post 1 day ago

Additional Jetson Orin Nano Super benchmarks will be added continuously.

posted an update 1 day ago

Post

368

Qwen3.5 on-device benchmarks on the Nvidia Jetson lineup are now live 🚀

We've added the latest Qwen3.5 models (0
8B - 9B) to our on-device inference benchmarks (Nvidia Jetson Orin Nano Super, AGX Orin, AGX Thor).

👉 Explore TPS, TTFT, E2E latency, and TPOT. Measured on real hardware: embedl/Edge-Inference-Benchmarks

🌟 Stay tuned for additional benchmarks and Embedl-optimized models: Enabling models run faster and on less expensive hardware.

If you're working on edge LLM deployment, we'd love to discuss your use case.

1 reply

·

New activity in embedl/Edge-Inference-Benchmarks 1 day ago

Add Qwen3.5 Benchmarks

#2 opened 2 days ago by

New activity in embedl/Edge-Inference-Benchmarks 2 days ago

Update chart and table styling

#1 opened 2 days ago by

updated 3 models 4 days ago

embedl/Cosmos-Reason2-2B-NVFP4A16

Image-Text-to-Text • 2B • Updated 4 days ago • 260 • 1

embedl/Cosmos-Reason2-2B-W4A16

Image-Text-to-Text • 2B • Updated 4 days ago • 8.76k • 6

embedl/Cosmos-Reason2-2B-W4A16-Edge2

Image-Text-to-Text • 2B • Updated 4 days ago • 11.9k • 10

liked a Space 4 days ago

Edge Inference Benchmarks

On-Device benchmarks across devices and models.

published a Space 4 days ago

Edge Inference Benchmarks

On-Device benchmarks across devices and models.

updated a dataset 4 days ago

embedl/documentation-images

Viewer • Updated 4 days ago • 5 • 2.91k

upvoted 2 collections 5 days ago

NVIDIA Jetson Orin Nano

Ultra-efficient model variants optimized for Jetson Orin Nano. Designed for constrained edge environments requiring low memory footprint. • 3 items • Updated 7 days ago • 2

NVIDIA Jetson AGX Orin

Models optimized and bench-marked for NVIDIA Jetson AGX Orin. Memory-efficient and latency-optimized variants designed for real-time edge inference. • 3 items • Updated 8 days ago • 2

published an article 5 days ago

Article

Benchmarks + Report: Optimized Cosmos-Reason2 (Qwen3-VL) for on-device inference on 8GB RAM (Jetson Orin Nano Super)

5 days ago

upvoted a collection 7 days ago

EdgeN

Quantization strategy where most weights are converted to INT4, activations remain in FP16, and sensitive layers are preserved in FP16. • 2 items • Updated 7 days ago • 1

upvoted a collection 8 days ago

FlashHead

Efficient Drop-In Replacement for the Classification Head in Language Model Inference. • 15 items • Updated 7 days ago • 1

updated a collection 8 days ago

FlashHead

Efficient Drop-In Replacement for the Classification Head in Language Model Inference. • 15 items • Updated 7 days ago • 1