Eldar Kurtic

ekurtic's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes

updated a model about 1 month ago

neuralmagic/Sparse-Llama-3.1-8B-2of4

Text Generation • Updated 6 days ago • 1.45k • 60

upvoted a paper about 2 months ago

"Give Me BF16 or Give Me Death"? Accuracy-Performance Trade-Offs in LLM Quantization

Paper • 2411.02355 • Published Nov 4 • 46

commented a paper about 2 months ago

"Give Me BF16 or Give Me Death"? Accuracy-Performance Trade-Offs in LLM Quantization

Paper • 2411.02355 • Published Nov 4 • 46 •

3

authored 4 papers about 2 months ago

Error Feedback Can Accurately Compress Preconditioners

Paper • 2306.06098 • Published Jun 9, 2023

MicroAdam: Accurate Adaptive Optimization with Low Space Overhead and Provable Convergence

Paper • 2405.15593 • Published May 24 • 1

Panza: A Personalized Text Writing Assistant via Data Playback and Local Fine-Tuning

Paper • 2407.10994 • Published Jun 24

"Give Me BF16 or Give Me Death"? Accuracy-Performance Trade-Offs in LLM Quantization

Paper • 2411.02355 • Published Nov 4 • 46

updated a Space about 2 months ago

Running

2

🔥

Quant Llms Text Generation

Quantized vs. Unquantized LLM: Text Generation Comparison

New activity in neuralmagic/Llama-3.1-Nemotron-70B-Instruct-HF-FP8-dynamic about 2 months ago

Context length

4

#2 opened about 2 months ago by

galigator

authored a paper 2 months ago

EvoPress: Towards Optimal Dynamic Model Compression via Evolutionary Search

Paper • 2410.14649 • Published Oct 18 • 8

updated 4 models 2 months ago

neuralmagic/Meta-Llama-3.1-8B-Instruct-FP8-dynamic

Text Generation • Updated Oct 19 • 1.67k • 5

neuralmagic/Meta-Llama-3.1-70B-Instruct-FP8-dynamic

Text Generation • Updated Oct 19 • 880 • 5

neuralmagic/Meta-Llama-3.1-405B-Instruct-FP8-dynamic

Text Generation • Updated Oct 19 • 270 • 14

daslab-testing/Llama-3.1-70B-Instruct-gptq4-128-True-seed1_mse1_staticFalse_clipFalse_fineweb

Updated Oct 18 • 3

updated a collection 2 months ago

FP8 LLMs for vLLM

Collection

Accurate FP8 quantized models by Neural Magic, ready for use with vLLM! • 44 items • Updated Oct 17 • 60

updated 2 datasets 2 months ago

neuralmagic/quantized-llama-3.1-leaderboard-v2-evals

Viewer • Updated Oct 10 • 247k • 498

neuralmagic/quantized-llama-3.1-humaneval-evals

Viewer • Updated Oct 10 • 73.8k • 57

updated a Space 2 months ago

No application file

🐢

Test

updated 2 models 3 months ago

daslab-testing/Meta-Llama-3.1-8B-Instruct-gptq4-128-True-seed1_mse1_staticFalse_clipTrue_fineweb

Updated Oct 3 • 46

daslab-testing/Meta-Llama-3.1-8B-Instruct-gptq4-128-True-seed1_mse1_staticTrue_clipTrue_orca_fineweb

Updated Oct 3 • 39