Hugging Face's logo

Avifenesh
/
memra-bench

GGUF
benchmark
cuda
speculative-decoding
nvfp4
Model card Files
xet
 Community
memra-bench

Commit History

readme: related links (engine repo + Hy3 overlay)
f10c3d0
verified

Avifenesh commited on

Upload README.md with huggingface_hub
a34f0cb
verified

Avifenesh commited on

Upload drafts/kf4/qwen36-35b-a3b-iq4xs/owngen-ranks-32768.txt with huggingface_hub
4b5804a
verified

Avifenesh commited on

Upload drafts/kf4/qwen36-27b-nvfp4/owngen-ranks-32768.txt with huggingface_hub
d821918
verified

Avifenesh commited on

Upload drafts/kf4/qwen35-9b-nvfp4/owngen-ranks-32768.txt with huggingface_hub
8b352e4
verified

Avifenesh commited on

Upload drafts/kf4/qwen36-35b-a3b-iq4xs/draft-owntrim-nvfp4head-q4blk.gguf with huggingface_hub
6556c0b
verified

Avifenesh commited on

Upload drafts/kf4/qwen36-27b-nvfp4/draft-owntrim-nvfp4head-q4blk.gguf with huggingface_hub
c9a0758
verified

Avifenesh commited on

Upload drafts/kf4/qwen35-9b-nvfp4/draft-owntrim-nvfp4head-q4blk.gguf with huggingface_hub
4c5494b
verified

Avifenesh commited on

Upload README.md with huggingface_hub
af287cb
verified

Avifenesh commited on

Upload README.md with huggingface_hub
01dab17
verified

Avifenesh commited on

Upload README.md with huggingface_hub
f2938f4
verified

Avifenesh commited on

Upload drafts/gemma4-e4b-qat/owngen-ranks-32768.txt with huggingface_hub
899f20a
verified

Avifenesh commited on

Upload drafts/gemma4-31b-qat/owngen-ranks-32768.txt with huggingface_hub
a07fb41
verified

Avifenesh commited on

Upload drafts/gemma4-26b-a4b-qat/owngen-ranks-32768.txt with huggingface_hub
de943a6
verified

Avifenesh commited on

Upload README.md with huggingface_hub
5b96aae
verified

Avifenesh commited on

Upload drafts/qwen36-27b-unsloth-nvfp4/owngen-ranks-32768.txt with huggingface_hub
e3809cc
verified

Avifenesh commited on

Upload drafts/qwen36-27b-unsloth-nvfp4/draft-owntrim-nvfp4head-q4blk.gguf with huggingface_hub
e193511
verified

Avifenesh commited on

Upload drafts/qwen36-35b-a3b-iq4xs/owngen-ranks-32768.txt with huggingface_hub
2f6f1d4
verified

Avifenesh commited on

Upload drafts/qwen36-35b-a3b-iq4xs/draft-owntrim-nvfp4head-q4blk.gguf with huggingface_hub
672bc27
verified

Avifenesh commited on

Upload drafts/qwen35-9b-nvfp4/owngen-ranks-32768.txt with huggingface_hub
1a444c9
verified

Avifenesh commited on

Upload drafts/qwen35-9b-nvfp4/draft-owntrim-nvfp4head-q4blk.gguf with huggingface_hub
133bb8b
verified

Avifenesh commited on

Upload drafts/qwen36-27b-nvfp4/owngen-ranks-32768.txt with huggingface_hub
f9e2c04
verified

Avifenesh commited on

Upload drafts/qwen36-27b-nvfp4/draft-owntrim-nvfp4head-q4blk.gguf with huggingface_hub
a47f0e8
verified

Avifenesh commited on

format decision: GGUF primary, ST best-effort
1ed6953
verified

Avifenesh commited on

27B ST serve config: + BW24_MMQ_F8F4=1 (f8f4 prefill tile adoption)
ec242fd
verified

Avifenesh commited on

Upload README.md with huggingface_hub
22a5457
verified

Avifenesh commited on

Upload prompts/p3-agentic-long-v2.txt with huggingface_hub
c73d86b
verified

Avifenesh commited on

Upload trims/frspec-corpus-32768.gguf with huggingface_hub
cba3c8e
verified

Avifenesh commited on

Upload trims/frspec-9bst-modelopt-32768.gguf with huggingface_hub
0e9ca61
verified

Avifenesh commited on

Upload CONFIGS.md with huggingface_hub
aed6470
verified

Avifenesh commited on

Upload README.md with huggingface_hub
19065da
verified

Avifenesh commited on

Upload README.md with huggingface_hub
5ed4420
verified

Avifenesh commited on

Upload README.md with huggingface_hub
028e450
verified

Avifenesh commited on

Upload README.md with huggingface_hub
a797b4d
verified

Avifenesh commited on

Upload README.md with huggingface_hub
67f6ade
verified

Avifenesh commited on

Upload README.md with huggingface_hub
61552dc
verified

Avifenesh commited on

Upload trims/frspec-9b-32768.gguf with huggingface_hub
bfbc697
verified

Avifenesh commited on

Upload folder using huggingface_hub
a9b430d
verified

Avifenesh commited on

initial commit
36af9e6
verified

Avifenesh commited on