4-5B Parameter LLMs — Page 38

affine-5CcXqzSZNHdcgg6ToWfQbjqDgMCNe1MDoydLKzaKFdxhXUHo

qwen3-4b-mini50

qwen3-4b-agentbench-exp03

Qwen3-4B-AgentBench-llm2025_advance_1st

manami-repo

Ewen3-4b-Instruct-2507

qwen3-4b-valid_solver_aux_v1

yuyangGPT

Affine-01-5EALnKDFv8qkqERMbTFoZWz2BBofuti1zRuvcRq1JCT81rdJ

affine-mine-5D4qsdevYnbVAgDDdKCkVpi36w14xMyGeQG5ijoNVmAW2ZNG

P9-split1_prob_Qwen3-4B-Base_0317-01

P2-split2_bs256_prob_Qwen3-4B-Base_0317-01

qwen3-4b-instruct-forc-rl

P2-split2_bs512_epoch5_5e-5_prob_Qwen3-4B-Base_0320-01

Qwen3-4B-CoderForge-SFT-baseline

SDRL-freq-Qwen3-4B-Base-majority_n8_l2048-GRPO_n8_bs256_long8-step200

P9-split3_prob_Qwen3-4B-Base_0322-01

qwen3_4b_baseline_solver_v3

qwen3_4b_baseline_solver_v5

Qwen3-4B-CoderForge-SFT-weighted-epoch3

qwen3_4b_baseline_v2_questioner_v5

qwen3_4b_vdrop75_v2_solver_v3

qwen3_4b_vdrop75_v2_solver_v4

qwen3_4b_vdrop85_solver_v1

4b-sft-qwen4b-step399

HiTOP-MedGemma4B-merged

qwen3_4b_sudoku_multi_act_rl_allow_one_action

GRPO_Qwen3-4B-Instruct-2507

MATH-TTT-Qwen3-4B-Base-Semantic-ClipHigh-Ent0.003-OpenAI

run1014-local-reasoning-baseline_lr1e-5_strict_F1_strictA2-step99

capybara-coder-4.1b

bank-model

affine-peg

qwen3_4b_instruct_sft

Affine-20251205-5232v2

qwen3-4b-thinking-rl-ckpt-109

qwen3-4b-thinking-rare-ckpt-109

qwen3_4b_easy_rl_final

qwen3_4b_medium_rl_final

qwen3_4b_sft_new

qwen3_4b_base_sft_final

affine-m-5