New AI Models (Last Year) — Page 96

paper_qwen_qwen3-instruct-4b_train_sft_all_train_dual

ds1p5b_code_sandbox-global_step_500

Qwen3-4B-CCC-irm-SafeRL

chess-special-80100

chess-special-85100

Advanced_Risk_Reward_Tampering_Qwen3-4B-Base

Qwen2.5-0.5B-Instruct-Gensyn-Swarm-tough_pesty_impala

affine-test-03

qwen3-4b-full-lora-step-180

ds-svd-muon-adam-1e-6-global_step_140

qwen3-1.7b-base-adam-1e-6-bs128-kl0.0-global_step_60

qwen3-1.7b-base-adam-1e-6-bs128-kl0.0-global_step_140

gemma-3-1b-it-gsm8k-structured-reasoning-grpo-stage-2-1

Llama-3.2-3B-Instruct_new_alpaca_003

qwen3_1.7b_rush_hour_one_move_4_9_epoch3

qwen3_1.7b_rush_hour_one_move_4_9

ds1p5b_code_sandbox-global_step_700

pdalma_ctx4_dm1_ce003_pr05_ptll32-1b_s2_ckpt_5_of_10_it36

ds1p5b_code_sandbox-global_step_800

qwen3_1.7b_rush_hour_multi_move_final_short_4_9_epoch1

qwen3_1.7b_rush_hour_multi_move_final_short_4_9_epoch2

Qwen3-4B-Exp-Stock

Llama-3.2-1B-Summarizer-merged

Qwen2.5-3B-Instruct_old_sft_alpaca_003

gemmasft

sft_qwen15_code200_lr_5e-6_constant_bsz_64_ckpt_5_of_5

qwen-arc-abs-gpt5.2-sft-fewshot4-1epoch-icmlpaper-0125

ds1p5b_no_if-global_step_700

Qwen3-4B-chess-grpo-base-5000

CR-CA

short_paper_qwen_2.json_train_dpo_v2_train_no_think

qwen3-0.6b-fine-tuned

pdcd200_cptq15_ce003_pr0_ptq25-15b_omi_c100k_200tok_s8_ckpt_4_of_10_it77

pdcd200_cptq15_ce003_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_4_of_10_it77

pdcd200_cptq15_ce003_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_6_of_10_it226

qwen3-1.7b-grpo-sft-base

pdcd200_cptq15_ce01_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_4_of_10_it77

pdcd200_cptq15_ce01_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_10_of_10_it1943

qwen_augment-inst

dpo-qwen-cot-merged

dpo-qwen-cot-merged

Qwen3-1.7B-CCC-merged-cp6-LR1e-4-irm