Yosef Worku Alemneh

rasyosef

AI & ML interests

Pretraining, Supervised Fine Tuning, Direct Preference Optimization, Retrieval Augmented Generation (RAG), Function Calling

Recent Activity

updated a model about 10 hours ago

rasyosef/Qwen3.5-2B-DSpark

updated a model 1 day ago

rasyosef/Llama-3.2-3B-Instruct-DSpark

updated a dataset 1 day ago

rasyosef/open-perfectblend-qwen-3.5-2b

View all activity

Organizations

updated a model about 10 hours ago

rasyosef/Qwen3.5-2B-DSpark

0.5B • Updated about 10 hours ago • 90

updated a model 1 day ago

rasyosef/Llama-3.2-3B-Instruct-DSpark

0.7B • Updated 1 day ago • 83 • 1

updated a dataset 1 day ago

rasyosef/open-perfectblend-qwen-3.5-2b

Viewer • Updated 1 day ago • 108k • 24

published a dataset 1 day ago

rasyosef/open-perfectblend-qwen-3.5-2b

Viewer • Updated 1 day ago • 108k • 24

updated a collection 1 day ago

DSpark & DFlash2

Collection

DSpark and DFlash2 draft models for faster inference with speculative decoding • 4 items • Updated 1 day ago

updated a model 1 day ago

rasyosef/gemma-4-E2B-it-dflash2

0.3B • Updated 1 day ago • 95

New activity in google/gemma-4-E2B-it-assistant 3 days ago

Up to 3.74x faster inference with a DFlash2 speculative decoding draft model for gemma-4-E2B-it (167 --> 626 tokens/s)

#2 opened 3 days ago by

rasyosef

New activity in google/gemma-4-E2B-it 3 days ago

3.7x faster inference with a DSpark (speculative decoding) draft model for gemma-4-E2B-it

#49 opened 5 days ago by

rasyosef

Up to 3.74x faster inference with DFlash2 speculative decoding draft model for gemma-4-E2B-it (167 --> 626 tokens/s)

#50 opened 3 days ago by

rasyosef

updated a model 3 days ago

rasyosef/gemma-4-E2B-it-dspark

0.3B • Updated 3 days ago • 67

New activity in unsloth/Llama-3.2-3B-Instruct 4 days ago

Up to 4x speedup with a DSpark speculative decoding draft model for Llama-3.2-3B-Instruct (35 tok/s --> 135tok/s on HumanEval )

#2 opened 4 days ago by

rasyosef

New activity in meta-llama/Llama-3.2-3B-Instruct 4 days ago

Up to 4x speedup with a DSpark speculative decoding draft model for Llama-3.2-3B-Instruct (35 tok/s --> 135tok/s on HumanEval )

#314 opened 4 days ago by

rasyosef

published a model 4 days ago

rasyosef/gemma-4-E2B-it-dflash2

0.3B • Updated 1 day ago • 95

published a model 6 days ago

rasyosef/gemma-4-E2B-it-dspark

0.3B • Updated 3 days ago • 67

published a model 9 days ago

rasyosef/Llama-3.2-3B-Instruct-DSpark

0.7B • Updated 1 day ago • 83 • 1

published a model 11 days ago

rasyosef/Qwen3.5-2B-DSpark

0.5B • Updated about 10 hours ago • 90

updated a model 12 days ago

rasyosef/Llama-3.2-1B-Instruct-DSpark

0.3B • Updated 12 days ago • 184

Yosef Worku Alemneh

AI & ML interests

Recent Activity

Organizations

rasyosef's activity

Up to 3.74x faster inference with a DFlash2 speculative decoding draft model for gemma-4-E2B-it (167 --> 626 tokens/s)

3.7x faster inference with a DSpark (speculative decoding) draft model for gemma-4-E2B-it

Up to 3.74x faster inference with DFlash2 speculative decoding draft model for gemma-4-E2B-it (167 --> 626 tokens/s)

Up to 4x speedup with a DSpark speculative decoding draft model for Llama-3.2-3B-Instruct (35 tok/s --> 135tok/s on HumanEval )

Up to 4x speedup with a DSpark speculative decoding draft model for Llama-3.2-3B-Instruct (35 tok/s --> 135tok/s on HumanEval )