Typed decision models for a laptop: Choice, Noul and Score with calibrated probabilities, one forward pass. Two sizes.
Ankit Aglawe
AnkitAI
rodoviario's profile picture
5H2lwhXaSqgOrVwtoarfKr7oH's profile picture
Galoserranoabad's profile picture·
AI & ML interests
Agentic fine-tunes · GGUF quantization · eval-gated releases
Recent Activity
updated a model 2 days ago
AnkitAI/TinyJev-4B updated a collection 2 days ago
TinyJev updated a collection 2 days ago
TinyJevOrganizations
Sensible defaults for everyday NLP: sentiment and emotion classifiers on modern architectures, published with their evaluation splits.
-
AnkitAI/Sensible-ModernBERT-Sentiment-Analysis
Text Classification • 0.1B • Updated • 42 • 1 -
AnkitAI/reviews-roberta-base-sentiment-analysis
Text Classification • 0.1B • Updated • 184 • 2 -
AnkitAI/deberta-v3-small-base-emotions-classifier
Text Classification • 0.1B • Updated • 443 • 2 -
AnkitAI/deberta-xlarge-base-emotions-classifier
Text Classification • 0.8B • Updated • 42 • 2
Open models trained on real Claude Fable 5 and GPT-5.5 agent traces. Eval-gated releases. Also on Ollama: ollama run parable/fable
-
AnkitAI/Parable-Qwen3-4B-Claude-Fable-5-GGUF
Text Generation • 4B • Updated • 408k • 19 -
AnkitAI/Parable-Qwen3-4B-Claude-Fable-5
Text Generation • 4B • Updated • 403 • 6 -
AnkitAI/fable-5-traces-sft-chatml
Viewer • Updated • 4.67k • 74 • 8 -
AnkitAI/Parable-Granite-4.1-8B-Claude-Fable-5-GGUF
Text Generation • 8B • Updated • 307k • 5
Financial sentiment models: fast DistilBERT and the ModernBERT flagship, benchmarked like-for-like against FinBERT.
Typed decision models for a laptop: Choice, Noul and Score with calibrated probabilities, one forward pass. Two sizes.
Open models trained on real Claude Fable 5 and GPT-5.5 agent traces. Eval-gated releases. Also on Ollama: ollama run parable/fable
-
AnkitAI/Parable-Qwen3-4B-Claude-Fable-5-GGUF
Text Generation • 4B • Updated • 408k • 19 -
AnkitAI/Parable-Qwen3-4B-Claude-Fable-5
Text Generation • 4B • Updated • 403 • 6 -
AnkitAI/fable-5-traces-sft-chatml
Viewer • Updated • 4.67k • 74 • 8 -
AnkitAI/Parable-Granite-4.1-8B-Claude-Fable-5-GGUF
Text Generation • 8B • Updated • 307k • 5
Sensible defaults for everyday NLP: sentiment and emotion classifiers on modern architectures, published with their evaluation splits.
-
AnkitAI/Sensible-ModernBERT-Sentiment-Analysis
Text Classification • 0.1B • Updated • 42 • 1 -
AnkitAI/reviews-roberta-base-sentiment-analysis
Text Classification • 0.1B • Updated • 184 • 2 -
AnkitAI/deberta-v3-small-base-emotions-classifier
Text Classification • 0.1B • Updated • 443 • 2 -
AnkitAI/deberta-xlarge-base-emotions-classifier
Text Classification • 0.8B • Updated • 42 • 2
Financial sentiment models: fast DistilBERT and the ModernBERT flagship, benchmarked like-for-like against FinBERT.