2-3B Parameter LLMs — Page 39

arete-llama-3.2-3b_5

qwen3-1.7b-dspo-no-sft-exp2

qwen3-1.7b-amr-20260206-1235

gemma

qwen3-1.7b-bilingual-amr-sft-v1

llm-vn-1-3b

llama_3.2_3b-owl_numbers_full_ep3

toolcalling-merged-demo

toolcalling-merged-demo

toolcalling-lora-demo

gemma_2b_it_Soccer

Llama-3.2-3B-Instruct-MPO-SKD-V7

pharos_2b

gemma2_2b-abstract-finetuned-ep1-b4

rankalign-v6-gemma-2-2b-d0.15-e2-hc-b2d-dbl-all-tcs-p0-nv1-ng1-fsx-sm0.1

rankalign-v6-gemma-2-2b-it-d0.15-e1-hc-b2d-dbl-all-fsx-sm0.1

Qwen3-1.7B-DeepMath-1024samples-RePO

DARE-MERGED

gemma-2-2b-it-star-10Rounds-iter-2

FL_gemma-2-2b-it-star-mixed_direct-OP-final_v2_10-2-3Rounds-iter-2

OrpoGemma2-2B

gemma-2-full-dare-peft

gemma2-2b-it-hh-grpo-harmless-step100

Mateno-v1.5-2b-law

gemmadpo7

gemma-2-2b-it_finetuned_4_new

gemma-2-2b-it_mtl

gemma-2-2b-it_finetuned_3_optimized1

gemma-2-2b-it-star-nl-OP_new_6epoch-final_v2_10-6-3Rounds-iter-2

cond_emotions_v2_ntr_25_nte_80_preamble_1proxy_20250525_233525

q4104

q448

easy-8k-med16k

hug10

hug5

hug6

llama3.2-3b-dpo-vanilla-OLD

EduRABSA_SLM_v1_SLERP_phi4mini

devspeedllm-ft-v0.0.6

Qwen3-1.7B

Qwen3-1.7B-GRPO-SRT-Math-12k-Stage-0

meta-llama_Llama-3.2-3B-Instruct-GRPO-vanilla_G_4