Vox MENS Qwen3 size ladder (brbrainerd)

Each ladder size has a Hub repo. A published rung is a Candle QLoRA adapter (candle_qlora_adapter.safetensors). Pull one with vox mens hub pull --size <size>. A not_trained row is the reserved repo, with no adapter yet.

Size Status Version Base Pin Repo Peak MB Final Loss Gate
0.6B diverged v2 Qwen/Qwen3-0.6B@c1899de289a04d12100db370d81485cdf75e47ca brbrainerd/vox-mens-qwen3-0.6b 15319 0.6898 ungated (collateral_damage_report.json (status != pass), gate_receipt.json (missing))
1.7B diverged v2 Qwen/Qwen3-1.7B@70d244cc86ccca08cf5af4e1e306ecf908b1ad5e brbrainerd/vox-mens-qwen3-1.7b 27007 0.6170 ungated (collateral_damage_report.json (status != pass), gate_receipt.json (missing))
4B diverged v2 Qwen/Qwen3-4B@1cfa9a7208912126459214e8b04321603b3df60c brbrainerd/vox-mens-qwen3-4b 34460 0.5435 ungated (gate_receipt.json (overall_passed != true))
8B in_sync v1 Qwen/Qwen3-8B@b968826d9c46dd6066d109eabc6255188de91218 brbrainerd/vox-mens-qwen3-8b 49374 0.5129 ungated (gate_receipt.json (overall_passed != true))
14B not_trained β€” Qwen/Qwen3-14B@40c069824f4251a91eefaf281ebe4c544efd3e18 brbrainerd/vox-mens-qwen3-14b β€” β€” β€”
32B not_trained β€” Qwen/Qwen3-32B@9216db5781bf21249d130ec9da846c4624c16137 brbrainerd/vox-mens-qwen3-32b β€” β€” β€”
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support