josefprusa/ThinkingCap-Qwen3.6-27B-int4-AutoRound-v1
Text Generation β’ 6B β’ Updated β’ 12.1k β’ 50
Maor PRO
Kroam1
AI & ML interests
Founder of StockAlloy.
Financial NLP, extraction, grounding.
Recent Activity
liked a model 19 days ago
josefprusa/ThinkingCap-Qwen3.6-27B-int4-AutoRound-v1 liked a dataset 19 days ago
StockAlloy/filedfact-passages posted an update 19 days ago
Structured financial facts, each linked to the character in the filing it came from π°π
Today I am introducing two datasets and a model trained on them:
https://huggingface.co/datasets/StockAlloy/filedfact-passages
A passage-complete dataset of phrases and the embedded financial facts.
https://huggingface.co/datasets/StockAlloy/filedfact-100k
A fact-sampled version- one grounded financial fact per row, deep-linked to its source.
https://huggingface.co/StockAlloy/filedfact-qwen3-8b
An 8B model fine-tuned to read a filing passage and return the facts as JSON.
https://huggingface.co/spaces/Kroam1/filedfact-demo
Live demo - paste a passage, see the facts.
Organizations
liked a
model 19 days ago
liked a
dataset 19 days ago
posted an update 19 days ago
Post
118
Structured financial facts, each linked to the character in the filing it came from π°π
Today I am introducing two datasets and a model trained on them:
StockAlloy/filedfact-passages
A passage-complete dataset of phrases and the embedded financial facts.
StockAlloy/filedfact-100k
A fact-sampled version- one grounded financial fact per row, deep-linked to its source.
StockAlloy/filedfact-qwen3-8b
An 8B model fine-tuned to read a filing passage and return the facts as JSON.
Kroam1/filedfact-demo
Live demo - paste a passage, see the facts.
Today I am introducing two datasets and a model trained on them:
StockAlloy/filedfact-passages
A passage-complete dataset of phrases and the embedded financial facts.
StockAlloy/filedfact-100k
A fact-sampled version- one grounded financial fact per row, deep-linked to its source.
StockAlloy/filedfact-qwen3-8b
An 8B model fine-tuned to read a filing passage and return the facts as JSON.
Kroam1/filedfact-demo
Live demo - paste a passage, see the facts.
- 1 reply
updated a
Space 19 days ago
updated a
model 19 days ago
updated 2
datasets 19 days ago
updated a
Space 19 days ago
reacted to DavidAU's post with β€οΈ 19 days ago
Post
20109
Qwen 3.6 27B Fine tune exceed 700 ARC-C for both 8 and 4 bit.
A new level of uncensored performance the puts this model squarely at "closed source" level of intelligence.
Model exceeds all critical benchmarks for both Qwen 3.6 27B AND Qwen 3.6 35B-A3B... and not by a little either.
Neo Imatrix MAX ggufs in both regular and MTP quants.
Benchmarks for Qwen 3.6 27B org and tuned, as well as Qwen 3.6 35B-A3B are up at the repo.
DavidAU/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-NEO-MAX-MTP-GGUF
A new level of uncensored performance the puts this model squarely at "closed source" level of intelligence.
Model exceeds all critical benchmarks for both Qwen 3.6 27B AND Qwen 3.6 35B-A3B... and not by a little either.
Neo Imatrix MAX ggufs in both regular and MTP quants.
Benchmarks for Qwen 3.6 27B org and tuned, as well as Qwen 3.6 35B-A3B are up at the repo.
DavidAU/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-NEO-MAX-MTP-GGUF
- 20 replies
published a
dataset 26 days ago
published a
Space 29 days ago
FiledFact-Qwen3-8B β research-preview SEC/XBRL fact extractor
#1 opened 29 days ago
by 
Kroam1
liked a
model 29 days ago
published a
model 29 days ago
published a
Space 30 days ago
liked a
dataset about 1 month ago
published a
dataset about 1 month ago