Models

Qwen3-4B-it-pira-ep3-qairm-ptbr

Qwen2-Math

merged_champion_v5_m3

BoyBarley-V28-Pro-Buddy

qwen25_7b_base_hc_sstt_n32_r1_dpo

MetalGPT-1

bug_fixing_rlvr-7b-nokl-v2

llama-3-8b-base-margin-dpo-hh-harmless-batch-size-64

nemosci-tasrep-a1mfc-dev1-maxeps-swes-r2eg__Qwen3-8B

OpenThinker-7B-reasoning-full-lora-max-type3-e1-2

nemosci-tasrep-a1mfc-gfistaqc-dev1-scaff-maxeps__Qwen3-8B

qwen3-8b-simnpo-gentle-baseline

qwen3-8b-4bit-sft-only-400-full-16bit

gemma-2-9b-it-lr3e-5-safedelta-scale0.5

gemma-2-9b-it-lr3e-5-safedelta-scale0.8

nepali_legal_qwen_merged_4

deepseek-r1-distill-qwen-1.5b-opencoder-educational-instruct-seed-42-G-4-merged

TinyLlama-1.1B-optimized

qwen3-8b-undial-baseline

Qwen3-0.6B-Farsi

Mlem-4B-SFT-Thinking-Seed2

tezos100k_continue_top8diverse100k_step2100__Qwen3-32B

tezos100k_continue_gptlongtezos_step900__Qwen3-32B

Qwen3-VL-8B-rslora-r32-2

GRPO-Think-1.5B-4k

Co-rewarding-I-Qwen3-8B-Base-DAPO14k

Megamind-v2-VL-high

Mlem-4B-RL-Thinking-Seed2

evolai-qwen3-1.7b-v1

Qwen3-8B-ep2_julia_codeforces_extended_with_thinksft_16bit_vllm

cliniq_model

DataMind-Analysis-Qwen2.5-7B

formai-tinyllama

qwen3-4b-instruct-medium2

qwen2.5-1.5b-loraplus-abstention

qwen2.5-0.5b-adalora-abstention

Qwen3-8B-PragReST-Vanilla-FullFT

e6172e5b

4e24b7ba

Gemma-3-4B-IT-ZH-SynthDolly-r16alpha128-E5-S73

sac-gspo-cl3e3-drgrpo-r1distill-qwen1.5b-step500-aime24-35-temp1

Llama-3.2-1B-Instruct-dpo