Collection of State-of-the-art FP8 Block Quantized Models
NM Testing
company
AI & ML interests
None defined yet.
Recent Activity
View all activity
models
552
nm-testing/Qwen3-VL-8B-Instruct-W4A16
3B
•
Updated
•
23
nm-testing/Qwen3-VL-8B-Instruct-NVFP4
6B
•
Updated
•
12
nm-testing/Qwen3-VL-4B-Instruct-NVFP4
3B
•
Updated
•
14
nm-testing/Llama-3.1-8B-Instruct-FP8-block
Text Generation
•
Updated
•
50
nm-testing/Llama-3.1-8B-Instruct-NVFP4-mse
5B
•
Updated
•
10
nm-testing/Llama-3.1-8B-Instruct-NVFP4-static_minmax
5B
•
Updated
•
9
nm-testing/TinyLlama-1.1B-Chat-v1.0-MXFP4
0.6B
•
Updated
•
8
nm-testing/EAGLE3-LLaMA3.1-Instruct-8B-sgl
Updated
•
34
nm-testing/Speculator-Qwen3-8B-Eagle3-converted-071-quantized-w4a16-sgl
Updated
•
11
nm-testing/Llama-3.2-1B-Instruct-attention-fp8-head
1B
•
Updated
•
6