2-3B Parameter LLMs — Page 24

qwen2.5-3b-dpo-mini

Qwen2.5-3B-Instruct_new_alpaca_005

SFT_DeepScaleR_Llama-3.2-3B_epoch_1_global_step_26

qwen3_1.7b_vanilla_romance_vanilla_ephishllm

Qwen3-1.7B-RFT-500

Llama-3.2-3B-Instruct-Base

qqWen-3B-Pretrain

qwen25_3b_qwen25_qwen3_rank_only-qwen25_qwen3_rank_only_cluster_4

qwen3b-security-audit

gemma2b-webxr-showroom-v2

qwen25_3b_instruct_qwen25_qwen3_rank_only-qwen25_qwen3_rank_only_cluster_3

llama_3.2_3b-owl_numbers_full_ep5

llama_3.2_3b-owl_numbers_full_ep9

armv8_to_riscv_qwen25coder_3p0b_full

toolcalling-merged-demo

qwen3_1.7b_summary_v10_vllm

gemma-2-2B-allenai-tulu-3-sft-math-MATH

6851_mcq_16_16_new_format_combined

gemma2-2b-it-hh-dpo-harmless-step-6000

mateno-v1

gemma-2-2b-it_RMU_s100_a300_layer3

SFT_peft

FFT_model_Gemma

gemma-2-2b-it_RMU_s400_a1200_layer3

6851_mcq_64_64

chatmodsft

chat1

Furina-experimental

6851_mcq_64_16_0318_sc

gemma2_dpo_oasst1_noise40_epoch2

gemma-2-2b-it_finetuned_1_optimized1

test

SFT_fft_Resta

magnet_mntp_msg-llama3.2-3b

llama3.2-3b-sft-3

gemma-empower-r16-inetune

Basqui-R1-4B-v1

gemma-2-it-tw

AutoRefine-Qwen2.5-3B-Instruct

ReasonFlux-Qwen3-dpo

reranker3b-sft

Qwen2.5-3B-RG-SFT