New AI Models (Last Year) — Page 130

qwen3_1.7b_easy_rl_ours_adv_fixed_gamma_995_98_ori_norm

final-vpt-gen_v2-8

llama3b-midtrain-open-thoughts114k_math-bs4-epoch1.0-ctx8192-ga1-lr1e-05-wr0.1-n4

final-joint_1-vpt-8

final-joint_2-vpt-8

sft_qwen32b

Qwen2.5-7B-Instruct-SFT-Pubmed-16bit-DFT

Vanilla_RL

affine-v124

qwen3_1.7b_easy_rl_ours_adv_fixed_gamma_1_98_gem_ms_seq_is

AT-qwen2.5-7b-hhrlhf-5120-sft-b3s3-tesla-ver10

zerp2

64b_SFT

meta-llama-Llama-3.1-8B-Instruct-sanitization-clean-OPI_SEP-42-202601102333

qwen3_1.7b_easy_rl_ours_adv_fixed_gamma_1_98_geo_ms_token_tis

16b_SFT

4b_SFT

gemma-3-27b-it-heretic

octothinker-hybrid-data_sft_50k_leon_nemotron_thinking-bs4-epoch1.0-ctx8192-ga1-lr5e-06-wr0.1-n4

jan13_8-8-1_sdf

qwen3_1.7b_easy_rl_ours_adv_fixed_geo_ms_only

qwen3_1.7b_easy_rl_final_group_norm

32b_SFT

affine-g15-5EhM3q9z5Yj4Vf2sgUSEbBTuqCvdMqQvFrnA3N9ZHnbxv7jG

2b_SFT_NEW

ee_lm8_grpo

qwen3_1.7b_easy_rl_ours_adv_fixed_geo_ms_seq_is

qwen3_1.7b_new_standard_A_sft_overfit_lr_5e_6__global_step_192

Vanilla_RL_NEW

qwen3_1.7b_new_standard_B_sft_overfit_lr_5e_6__global_step_396

qwen3_1.7b_new_standard_B_sft_overfit_lr_5e_6__global_step_792

qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_5__global_step_888

qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_5__global_step_592

qwen3_1.7b_new_standard_A_sft_overfit_lr_5e_6__global_step_288

qwen3_1.7b_new_standard_A_sft_overfit_lr_5e_6__global_step_384

qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_6__global_step_1480

qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_6__global_step_1184

qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_6__global_step_296

qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_6__global_step_592

self-debate-exp-Qwen3-4B-Base-majority_n4_l2048-DAPO_n8_bs256_long8-step200

qwen317step300

64b_RL_DAPO