New AI Models (Last Year) — Page 528

KG-R1-CWQ-hit1-no-turn-advantage

gemma-2-9b-it-gsm8k-sn-tuned-lr3e-5

llama2_7b_chat-WaRP-original-space-gsm8k-lr5e-5

Qwen3-0.6B-OURS_self-g_general_reward_e_sycophancy_keep_last-100-tokens_w3-seed_0

qwen2.5-3b-avap-v3c

qwen3vl-pid-sft-full-1500

qwen3vl-flowchart-to-mermaid

tezos100k_continue_tezos_step1200__Qwen3-32B

Qwen3-VL-8B_reasoning_answer_v3_3epoch

llama3.1-8b-base-lr1e-5-gsm8k-safedelta-scale0.1

Qwen_plus2_shot7_sft_fold0

unsup-Llama-3.1-8B-Instruct-datav2-only_mask_w_item_mesh

llama-2-13b-chat-hf-only-sn-tuned-lr5e-5

qwen_4b_with_all_data_v1.0.0_epoch_3

qwen2-7b-rag-ko-checkpoint-813

d1_harden_then_constrain_top4_seq_glm47

d1_hardened_top4_seq_glm47

pakistan-leaders-tinyllama-peft-merged

Qwen3-4B-hydro-sft

qwen3vl_ins_math_10k

math_m32-4b-9e032637-not_easy_1e-4_800

seed0_sample5000_bmlama_google-gemma-3-4b-it_en-fa_1.0-1.0_1.0

gemma-3-4b-it-128k-presls

smartclaims-grpo-unk10

qwen2.5_math_1.5b_grpo_aspo_rollout_8

gemma-2-9b-it-lr3e-5-safeinstr-0.05

qwen3vl_think_math_10k

llama3.1_8b_instruct-Safety-FT-lr3e-5

Qwen2.5-Coder-TA-MCEVALHARD-1.5B-Base

seed0_sample5000_bmlama_meta-llama-Llama-3.1-8B-Instruct_en-fa_1.0-1.0_1.0

Edu-OPCD-train16-k10-lr5e-7-ema0.01-eopd0.8-qwen3-4b-think-edu_merged_insensitive20

llama3.1_8b_sft-solo-bos-attn-k28

llama3_2_3b_instruct_only_rsn_tuned_lr5e-5

evolai-qwen2.5-1.5b-sn47-v2

gemma-2-9b-it-lr3e-5-gsm8k-lr1e-5

seed0_sample5000_bmlama_meta-llama-Llama-3.1-8B-Instruct_en-fa_DPO_5e-06

diallm-gemma-sft-aus

Gemma-3-4B-IT-HI-SynthDolly-1A-E3

llama2_7b_chat-gsm8k_FT_lr3e-5

1B-Instruct-Tulu-full

unsup-Qwen3-8B-datav3-only_mask_w_item_mesh

Qwen3-8B-slimllm-3bit-calibration-Chinese-128samples