New AI Models (Last Year) — Page 640

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

gcjg134f

llama2_7b_gsm8k_ft_freeze_sn_lr3e-5

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

llama3_2_3b_instruct_only_rsn_tuned_lr5e-5

Minmax-TOFU-2

Affine-95-5GC6UdKaWXUoY9a9RVcGusCQ1J8tKDyE4Kv8FMzdMoBN4RHx

llama2_7b_chat-gsm8k_FT_lr3e-5

ours_gemma_1b_output_dist_merged

llama3.1_8b_instruct_MATH-FT-resta-gamma0.3-lr5e-5

llama3.1_8b_instruct_MATH-FT-lr3e-5

lorem_advshape_qwen3-1.7b-base

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

llama3_2_3b_instruct_only_sn_tuned_lr5e-5

On-policy-GRPO

llama3_8b_instruct-MATH_FT_lr5e-5

Qwen3-8B-Base-baseline-ghpo

bs16-k10-lr5e-7-ema0.01-eopd0.8-qwen3-4b-think-sciknoweval_material_pos_sens_bottom20

Fanar-1-9B-SFT-safe

Qwen2.5-Coder-LEAK-MCEVALHARD-1.5B-Base-7

Qwen2.5-Coder-LEAK-MCEVALHARD-1.5B-Base-9

tinyllama-1.1b-lora-risk-classifier-v1

qwen3-4b-medrect-assessor

phi2-docstring-model1

llama-3.1-8b-instruct-math-rsn-tuned-lr5e-5

Qwen3-4B-Instruct-2507-ScaleSWE-Distilled-Epoch1

gptlong_openthoughts3_smoke__Qwen3-32B

qwen3_8b_lora_query_planner

llama3.2-1b-Inst-safegrad

tezos100k_continue_tezos_step2700__Qwen3-32B

ddp-llama32-1b-ultrachat

Qwen3-4B-Instruct-2507-ScaleSWE-Distilled-Epoch3

g1_top8_85k_gptlong_swegym_32b_step4425__Qwen3-32B

tezos100k_continue_top8diverse100k_step3900__Qwen3-32B

tezos100k_continue_tezos_step2400__Qwen3-32B

gptlong_continue_gptlongtezos_step3300__Qwen3-32B

tezos100k_continue_top8diverse100k_step4200__Qwen3-32B

tezos100k_continue_gptlongtezos_step2400__Qwen3-32B

wos-main-qwen35

wos-meeting

Qwen_Qwen3-4B-Thinking-2507_fp3-e1m1_openr1-default-concat_2048_8_1024_256_lr0.03

Qwen_Qwen3-4B-Thinking-2507_int3-g128_qwen3-traces-cot-concat_2048_8_1024_256_lr0.03