Open Models, Ready for Production

slm-agent-evidence-judge-smollm2-135m

slm-agent-action-router-qwen2-5-1-5b

slm-agent-evidence-judge-qwen2-5-0-5b

slm-agent-action-router-smollm2-135m

slm-agent-action-router-qwen2-5-0-5b

slm-agent-evidence-judge-smollm2-360m

holo31-mixed-nvfp4-bf16-head-overlay

axon-oss

axon-250m

Shaer-adapters-grpo-short1k-no-trio-v4

tomaszki

model-long-acc-1

Karroyan

MasterMInd-Dou-full-kl

llama3-8b-instruct-math-ties-k70

llama3-8b-instruct-math-ties-k90

llama3-8b-instruct-math-ties-k50

llama3-8b-instruct-math-ties-k30

llama3-8b-med-math-ties-k90

llama3-8b-instruct-math-ties-k10

tomaszki

model-long-acc-4

llama3-8b-med-math-ties-k50

Qwen3.5-9B-rl-tree-dapo

llama3-8b-med-math-ties-k70

firzahdzm

1gpu-dpo-d8589b56-denfix1

BrickNet-1.7B-PT

qlora_50

llama3-8b-med-math-ties-k10

qwen3.5-0.8b-cost-aware-router

llama3-8b-med-math-ties-k30

tomaszki

model-long-acc-2

amityco

tau-max-retail-next-action-manual-v0.6-sft-high-dpo-high-0.6b-sft-reject

eekay

Llama-3.1-8B-Instruct-noised-np0.15-emb-s0

emotional-dialogpt-small

emotional-dialogpt-medium

ApocalypseParty

G4-31B-configHA

emotional-dialogpt-large

llama3-8b-instruct-med-ties-k70

llama3-8b-instruct-med-ties-k90

emotional-gpt2

emotional-distilgpt2

llama3-8b-instruct-med-ties-k50

Jordine

cadenza-echoblast-sdf-v3redo-iter2a-qwen35-27b-v1

Jordine

cadenza-echoblast-denial-iter2a-balanced-qwen35-27b