A Jev-style decision model was created by fine-tuning Qwen3.5-4B on a typed decisions dataset using 4-bit QLoRA training on a single A100 GPU in 15 minutes, achieving 80% accuracy. The model adds a decision head to classify decision types from text inputs.

Name: A Jev-style decision model was created by fine-tuning Qwen3.5-4B on a typed decisions dataset using 4-bit QLoRA training on a single A100 GPU in 15 minutes, achieving 80% accuracy. The model adds a decision head to classify decision types from text inputs.
Uploaded: 2026-10-17T03:00:33.891Z
Channel: analogalok
Description: A Jev-style decision model was created by fine-tuning Qwen3.5-4B on a typed decisions dataset using 4-bit QLoRA training on a single A100 GPU in 15 minutes, achieving 80% accuracy. The model adds a decision head to classify decision types from text inputs.

analogalokagent-decisions11.5K views · 118 likes

Discovered 2026-10-17, refreshed 2026-10-18. Descriptions and stats are pulled from the project's own X post and refreshed automatically — they aren't independently verified by JevTracks beyond the initial eligibility check.