Open Models, Ready for Production

first_finetune

tarona

qwen3-4b-sft-adapter_v1_001

qwen3-4b-structured-output-lora2ep1e4005datagrade

tussiiiii

qwen3-4b-structured-output-lora-continued-v5-ver2

hkust-nlp

drkernel-14b

hkust-nlp

drkernel-8b

flux_40_50_sky_wandb_yhn_adamw_lr8e4_LORA16

starter_sft_0001_lr1e-6

TrustNet

GLM-4.7-Flash-FP8

IntentNet

Adanato

Meta-Llama-3-8B-Instruct_qwen3_gemma_rank_diff-qwen3_gemma_rank_diff_cluster_1

qwen3-4b-structeval-v4-2epoch

qwen3-4b-structured-output-lora-v5

gemma_3_4b_it-lora-r8-eng-hau

gemma_3_4b_it-lora-r8-amh-eng

qwen3-4b-structured-output-lora

Adanato

Meta-Llama-3-8B-Instruct_qwen25_gemma_rank_diff-qwen25_gemma_rank_diff_cluster_1

qwen3-4b-structured-output-lora-v4

tarona

qwen3-4b-dpo-marged_001

gemma_3_4b_it-lora-r8-eng-amh

LLM_main_003_DPO

KKanno

qwen3-4b-structeval-sft-lora-3epoch

Qwen3-VL-4B-Instruct-v6_s5_exp1_only_blocks-full_SFT

Qwen3-VL-4B-Instruct-v6_s4_exp2_only_blocks-full_SFT

Qwen3-VL-4B-Instruct-v6_s4_only_blocks-full_SFT

Qwen3-VL-4B-Instruct-v6_s5_only_blocks-full_SFT

Qwen2.5-VL-7B-Instruct-obj_4k-full_SFT

qwen3-4b-structured-output-lora-rev.08

Qwen3-VL-4B-Instruct-v6_s2_exp_only_blocks-full_SFT

Qwen3-VL-4B-Instruct-v6_s1_s1_exp_only_blocks-full_SFT

Qwen3-VL-4B-Instruct-v6_s2_only_blocks-full_SFT

jasonrb

qwen25-coder-32b-pangolin-merged

fumikawa

a25-v0008

Qwen2.5-VL-7B-Instruct-scene_4k-full_SFT

dpo-qwen-cot-merged

Qwen3-VL-4B-Instruct-scene_4k-full_SFT

Qwen3-VL-4B-Instruct-mc_ex_cs-full_SFT

Qwen2.5-VL-3B-Instruct-v6_s5_only_blocks-full_SFT

Affine-GregNew

Qwen2.5-VL-7B-Instruct-mc_cs-full_SFT

Qwen3-VL-4B-Instruct-mc_cs-full_SFT